- Inkscape Devel - Inkscape Mailman

Announcing Inkscape 0.37 Release
by Bryce Harrington 16 Feb '04

16 Feb '04

A new version of Inkscape is now available for use. Inkscape 0.37 includes many major new features, numerous bug fixes, and extensive codebase cleanup. Briefly ======= * New functionality: boolean operations on paths, dynamic and linked offsets, path outlining, standalone SVG viewer, drag-and-drop, letter-spacing. * More convenient editing: top panel, new color widget, 256x zoom, snapped rotation with Ctrl everywhere, selecting within groups, lots of new shortcuts, and more. * Interface and usability: saving view with the document, zoom history, fullscreen, hide/show all open dialogs, smarter export dialog, helpful statusbar, and much more. * Inkscape internals: built with C++, brand new classes for points, rects, matrices, and URIs, massive cleanups. * Last but not least: tutorial for new users, keys and mouse reference, Inkscape User mailing list, hassle-free Windows builds. More Detailed Info ================== Release Notes http://www.inkscape.org/cgi-bin/wiki.pl?ReleaseNotes037 Screenshots http://www.inkscape.org/screenshots/index.php Download From ============= http://sourceforge.net/project/showfiles.php?group_id=93438 RPM, Debian, Windows, and source versions are available. Please visit http://www.inkscape.org to learn more, join our User Mailing List, chat with the developers, and submit your ideas for making Inkscape the leading SVG-compliant vector graphics tool.

1 0

Strange bug in Windows build?
by "Áki G. Karlsson" 16 Feb '04

16 Feb '04

Hi I cannot keep my mouth shut anymore about this. It is probably some problem with my system if nobody else has noticed it ( I could not find anything similar in the buglist ). I am using WinXP (Home) and have always had this same problem with Inkscape, right from the first release. It has thus been unusable for me at work ( I have it on a linux box at home and it works fine there ). The most recent build I have is Bob Jamison's windows build called Inkscape 0.37-20040210. I have several of the GNU libraries installed in PATH, but none that I know of in system32. The thing is that if I use the freehand tool ( or any other tool than the predefined shapes ), as soon as I release the mouse button the shape literally goes haywire, that is, it becomes a big spindly mess. I set up four jpegs to show this effect. Here's one showing drawing with the freehand tool ( http://www.hi.is/~akig/inkscape1.jpg ) And here's after I release the mouse button ( http://www.hi.is/~akig/inkscape2.jpg ) Here's one showing a star *after* I've released the mouse button ( http://www.hi.is/~akig/inkscape3.jpg ) And here's what happens if I convert it to a shape... ( http://www.hi.is/~akig/inkscape4.jpg ) :( Hope this is just me and nothing serious... Best regards Áki

2 1

Mandrake RPM package (preliminary)
by Bryce Harrington 16 Feb '04

16 Feb '04

I've done a first cut at a Mandrake RPM package. There's nothing Mandrake-specific except the name, so in theory this would work for any RPM-based system. http://bryceharrington.com/inkscape-0.37-1.mdk90.i586.rpm Note that this is an i586 arch build instead of i386 or i686 as planned; I'm building the latter two presently and will upload them when they're done. The thing I need some feedback on is that I specified the Requires parameter listing out what appear to be the dependent libs & etc. I need to know if you try this rpm on your system and it reports that it won't install due to dependency issues. (The adventurous are encouraged to try --force and --nodeps to see if the rpm will work anyway.) Here's an SRPM, too: http://bryceharrington.com/inkscape-0.37-1.src.rpm Bryce

3 7

libnrtype -> Pango replacement plan
by Bryce Harrington 16 Feb '04

16 Feb '04

One of the tasks for this release was to investigate what tasks would be needed to replace libnrtype with Pango for font handling. I did a bit of looking around to see where libnrtype functions are being called. Here's my first cut at what it looks like might need to be done. Please review and comment on any additional tasks that you think will need to be done: * Add Pango as an optional dependency for Inkscape & test on our various target platforms * Modify widgets/font-selector.cpp to use Pango font new/ref/deref function as a compile-time option * Modify dialogs/text-edit.cpp to use Pango font new/ref/deref function as a compile-time option * Modify display/nr-arena-glyphs to use Pango font functions as a compile-time option * Create replacements for sp-text.cpp and sp-chars.cpp that uses Pango instead of libnrtype as compile time option * Test removal of libnrtype from Inkscape's Makefile & see what it breaks * Make Pango a required dependency for Inkscape & remove compile-time option scaffolding I've added these tasks to 0.43 and made Pangoification the objective for that release. If someone gets around to doing the tasks before that, that'd be lovely, otherwise we can plan on tackling it then. Bryce

2 1

Call For Clipartists
by Bryce Harrington 16 Feb '04

16 Feb '04

With Myrk's nifty ribbon.svg submission and Simarilius' SVG Signs contribution this morning, we're now officially ready to receive SVG clipart collections for Inkscape. Clipart is extremely important for users of vector art programs, because it can save a lot of time when making certain kinds of drawings, and to provide examples users can start with when making their own creations. I think it'd be most effective (and fun) to leave this process fairly open ended and flexible, since that's how we like to do things in this project. What we plan to do is treat each clipart "collection" as it's own discrete package, that is welcome to set its own rules and conventions as its contributors see fit. We'll distribute a few with Inkscape itself, others can be provided as separate add-ins. So, if you want you can establish your own collection of just your own work, or can collaborate with other users for a group effort. You can set up some organization and standards for quality control or to maintain a theme, or leave it freeform. Whatever works best for you. Let's collect info and ideas about Clipart Collections on this wiki page, to document HOWTO's, ideas, questions, guidelines, etc.: http://www.inkscape.org/cgi-bin/wiki.pl?ClipartContribution If you want to start working on a clipart collection, please feel free to add a link to the appropriate page(s) there. Bryce

2 1

Re: [Inkscape-devel] ru.po
by bulia byak 15 Feb '04

15 Feb '04

>>http://inkscape.org/cgi-bin/wiki.pl?RussianTerminology >Mozilla defaults to japanese for that link - is there a way to force >KOI8-R, or perhaps unicode? Just use View > Encoding > Auto-detect > Rusian and Mozilla will currectly show pages in any Russian encodings (provided you have Cyrillic fonts). Of course it would be nice to specify correct encoding for that page explicitly, but I don't know how to do that for one Wiki page (without affecting other pages). _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

RE: [Inkscape-devel] ru.po
by bulia byak 15 Feb '04

15 Feb '04

>Here is russian messages based on inkscape.pot of 0.37 release and messages >from sodipodi 0.33 translation. Thanks! I edited it a bit, fixing some typos and shortening awkwardly long sentences. I also changed a few badly translated terms and started a wiki page for Inkscape's Russian terminology: http://inkscape.org/cgi-bin/wiki.pl?RussianTerminology You are welcome to discuss by editing and expanding that page. Please stick to these terms when translating the tutorial. _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

2 1

guides toggle and other fixes
by bulia byak 15 Feb '04

15 Feb '04

Now: - The | key (Shift+\) toggles guide visibility and snapping to guides (if you need to set them separately, use Document Options dialog) - The # key (Shift+3) toggles grid visibility and snapping to grid (if you need to set them separately, use Document Options dialog) - Both these settings are saved with the document - When a new guide is being created by dragging from the ruler, guide visibility and snapping is switched on - Default showguides (if not set) is on, for backward compatibility - The Document Options dialog now always reflects the current state of these (and other) parameters, so you can press these keys (or edit your namedview in XML editor) and watch the checkbuttons toggle _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=dept/features&pgmarket=en-ca&RU=http%3a%2f%2fjoin…

2 1

Re: [Inkscape-devel] bug: win32: Saving to wrong location.
by bulia byak 15 Feb '04

15 Feb '04

>I will look at that one. Looks like it >might be a general bug, not just Win32. I tried to reproduce it in Linux today but couldn't - it works OK here. And I really don't see where in the code might that happen. Bob: saving uses document->uri, which is set by sp_document_create and sp_document_set_uri, but I have no idea how importing can affect it. _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

1 0

bug: win32: Saving to wrong location.
by John Cliff 14 Feb '04

14 Feb '04

Hi guys, Found another one, If you save an svg somewhere then import an image from somewhere else when you do save at a later point it saves the file to the location you imported from not the location you saved it too. opened a tracker item on it, #897117. cheers John __________________________________ Do you Yahoo!? Yahoo! Finance: Get your refund fast by filing online. http://taxes.yahoo.com/filing.html

2 1

RE: [Inkscape-devel] offset path inaccuracies
by bulia byak 14 Feb '04

14 Feb '04

>I was playing with the linked offset on some very simple paths and found it >didn't create a smooth path: > >http://members.rogers.com/rcrosbie/svg/king_solomon_knot.jpg >http://members.rogers.com/rcrosbie/svg/king_solomon_knot.svg It's a known bug in offset. Nathan says he may be able to fix it sometime. As a workaround, enlarge the object several times, do the offset, and then reduce it back - this way the distortions are relatively smaller. _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=dept/features&pgmarket=en-ca&RU=http%3a%2f%2fjoin…

1 0

offset path inaccuracies
by Robert Crosbie 14 Feb '04

14 Feb '04

Hello, I was playing with the linked offset on some very simple paths and found it didn't create a smooth path: http://members.rogers.com/rcrosbie/svg/king_solomon_knot.jpg http://members.rogers.com/rcrosbie/svg/king_solomon_knot.svg Are there inaccuracies in the linked offset algorithm, or is it something to do with the paths and the values I was trying? I find similar things happen with the dynamic offset as well, using the same values, but it's not as obvious. This was from an svg file I saved 3 years ago using Jasc Web Draw. Great job with the 0.37 release BTW. Cheers, Rob. http://members.rogers.com/rcrosbie _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

1 0

RE: [Inkscape-devel] Some fairly random feature reqs + bug reports.
by bulia byak 14 Feb '04

14 Feb '04

> > >see > > >http://members.lycos.co.uk/simarilius/weirdness.jpg I think I know what the problem here is. Johan, in verbs.h you have: #ifdef HAVE_GTK_WINDOW_FULLSCREEN SP_VERB_FULLSCREEN, #endif /* HAVE_GTK_WINDOW_FULLSCREEN */ /* Dialogs */ SP_VERB_DIALOG_DISPLAY, SP_VERB_DIALOG_NAMEDVIEW, SP_VERB_DIALOG_TOOL_OPTIONS, SP_VERB_DIALOG_FILL_STROKE, SP_VERB_DIALOG_TRANSFORM, SP_VERB_DIALOG_ALIGN_DISTRIBUTE, SP_VERB_DIALOG_TEXT, SP_VERB_DIALOG_XML_EDITOR, SP_VERB_DIALOG_TOGGLE, SP_VERB_DIALOG_ITEM, /* Footer */ SP_VERB_LAST Likely, the value of HAVE_GTK_WINDOW_FULLSCREEN is for some reason different here than in the rest of the codebase (not included config.h?) and therefore the constants for the dialog verbs are all shifted by one: transform becomes fill&stroke, etc. This explains the confused shortcuts. I'm not sure why this happens only on Windows (or perhaps it happens when HAVE_GTK_WINDOW_FULLSCREEN is false?) Could you please debug this ASAP, as it's a pretty annoying issue. Please call ishmal for help with Windows setup. _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 3

RE: [Inkscape-devel] bug with moving selections
by bulia byak 13 Feb '04

13 Feb '04

>interesting thing is sodipodi doesnt do it... Interesting indeed! Could you please test other mousedrag operations, such as panning canvas, freehand drawing, creating shapes etc.? Do they all suffer from this as do the selector operations? >import is working on all three of those formats on >windows 2k currently, cant speak for other windows >versions. Thanks, closing. _________________________________________________________________ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

3 3

Re: [Inkscape-devel] Manpage for inkview
by bulia byak 13 Feb '04

13 Feb '04

>build different formats. I wonder if docbook can convert our tutorial.svg >to TeX ? No, because svg is not semantic, it's a low-level visual representation. I can elevate the content of the tutorial into DocBook or other semantic XML, but I see no benefit to it, because the tutorial is interactive - it teaches you to use Inkscape from within Inkscape, and thus contains instructions like "move this object with selector" which would make no sense in any format other than Inkscape SVG. _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 3

Manpage for inkview
by Wolfram Quester 13 Feb '04

13 Feb '04

Hi alltogether! While building the debian package, lintian told me binary-without-manpage inkview So yesterday in the evening I converted http://www.inkscape.org/cgi-bin/wiki.pl?Inkview into a small manpage to be included in the debian distro. It is attached, so you can use it if you want and change it or tell what to do better. Best regards, Wolfi

3 4

RE: [Inkscape-devel] printing + gradient
by bulia byak 13 Feb '04

13 Feb '04

>well.. just a 2 cents (euro) stupid idea/question >i export from inkscape to png a gradient >i import that png into ooffice >i print the document (via cups on redhat )? postcript ? >the gradient print well... If you already have bitmap (PNG) created by Inkscape, it's easy to print it from anywhere. It can even be embedded into Postscript (still remaining a bitmap). The problem is that Postscript that is directly generated by Inkscape does not embed bitmaps for gradients (or transparency) and thus distorts them. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

printing + gradient
by herve couvelard 13 Feb '04

13 Feb '04

bulia byak wrote: > My guess is that you print via Postscript, and Postscript does not > support gradients. When other vector apps export things like > transparency or gradients to PS, they "simulate" them, usually by > bitmaps. Inkscape does not yet do that. But it's just a guess, your bug > may be different. Please do file a bug report. well.. just a 2 cents (euro) stupid idea/question i export from inkscape to png a gradient i import that png into ooffice i print the document (via cups on redhat )? postcript ? the gradient print well... ooffice did the bitmap conversion ? cups don't use postcript ? i understood nothing ? Hervé

1 0

bug with moving selections
by John Cliff 13 Feb '04

13 Feb '04

Hi folks, Found a bug that seems a bit weird, If you create a shape, doesnt matter what, go to select and drag it, but move the mouse outside of the drawing area, it becomes impossible to interact with the interface again until you do another select action. Ie if you try to change tool, use a menu, close the program etc, it behaves like its hung. If the shape is the only thing you've got in the document the only way to get control back is to click on it again, made me think the whole thing had died for a minute there. Have only tried it on Win 2k, but it does it in both the builds I've got on my home box, and the 2 on my work one. cheers John btw isnt bug 894662 closed? __________________________________ Do you Yahoo!? Yahoo! Finance: Get your refund fast by filing online. http://taxes.yahoo.com/filing.html

2 1

RE: [Inkscape-devel] bug with moving selections
by bulia byak 13 Feb '04

13 Feb '04

> Found a bug that seems a bit weird, >If you create a shape, doesnt matter what, go to >select and drag it, but move the mouse >outside of the drawing area, it becomes impossible to >interact with the interface again until you do another >select action. Ie if you try to change tool, use a >menu, close the program etc, it behaves like its hung. >If the shape is the only thing you've got in the >document the only way to get control back is to click >on it again, made me think the whole thing had died >for a minute there. That seems to be http://sourceforge.net/tracker/index.php?func=detail&aid=852976&group_id=93… Obviously the reason is that on Windows, GTK sends different (or no) events when mouse is outside the window. This needs investigation, unfortunately we have only one developer actively working on the Windows version. >btw isnt bug 894662 closed? I guess it waits for confirmation. If you can confirm that you can import PNG, GIF, JPEG with the latest win32 build, I'll close it. _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

RE: [Inkscape-devel] fr.po (again)
by bulia byak 13 Feb '04

13 Feb '04

>I join an update fr.po . Merci beacoup. It's uploaded to CVS. Now, if you can compile the latest CVS with French interface, it would be nice if you could send us a screenshot (with some dialogs, menus etc.) for the site. _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

3 2

ru.po
by Alexey Remizov 13 Feb '04

13 Feb '04

Hello. Here is russian messages based on inkscape.pot of 0.37 release and messages from sodipodi 0.33 translation. 535 translated messages, 8 fuzzy translations, 57 untranslated messages. -- С уважением. | WBR. | Алексей. | Alexey. | ----------------------------| mailto:alexey@...188... | jabber:remizov@...189... | ----------------------------'

1 0

PROPOSAL: Reorganizing Directories
by Bryce Harrington 13 Feb '04

13 Feb '04

I'd like to propose some structural changes to both the codebase directory hierarchy and the directories created during installation. If these changes are acceptable, they could be introduced for the 0.38 development cycle. Bryce Distribution / Packaging Files ============================== Presently, we have a number of files in the codebase related to generation of distribution packages. Some of these files are in the root inkscape/ directory, others are in subdirs like inkscape/debian/. To head off potential clutter as Inkscape becomes supported on a wider variety of systems, we should collect as many of these packaging files as feasible into a subdir called inkscape/packaging/. Inside that directory would be distro-specific subdirs, plus a common directory for shared scripts and collateral: inkscape/packaging/ common/ debian/ fedora/ fink/ mandrake/ suse/ Config Files ============ Currently we do not have global config files, but instead put the config file in the user's home dir in ~/.inkscape/. These are generated at run-time by Inkscape itself, so the defaults are hard-coded into the application, and can't be customized by the sysadmin except by editing the appropriate header file and recompiling/repackaging the app. The correct approach for Linux apps is to have both user and global configs, and to use the global as the default if the user does not yet have a config file. This way, site customizations are known (such as the fact that its shared files are in /usr/share, vs. /usr/local/share, vs. /opt/inkscape/share, etc.) The proposal is for Inkscape to establish its config file location in /etc/inkscape/, and to place a global preferences.xml file there, as well as any other config files we need later, like extensions.xml, etc. Ideally, the global config file would be loaded first, and then the user config file "override" any portion of the global options. This way the user config file only needs to contain customizations, so that if we choose to change the defaults in a future install, the user's config files do not necessarily become invalid. Inside the CVS codebase, the default etc. files would be stored in inkscape/etc/, instead of being produced by the application at runtime. During installation, installation-specific parameters (such as alternate paths for various files) will be substituted in appropriately. "Share" Collateral ================== Inkscape currently generates a /usr/local/share/inkscape/ directory during install, and copies the contents of the codebase's icons/ directory to that location. We need more pieces of collateral installed, and need to do so in a more organized manner. The recommendation is to establish a set of subdirs in /usr/local/share/inkscape/, as follows: AUTHORS NEWS clipart/ examples/ extensions/ fonts/ gradients/ icons/ keyboards/ markers/ palettes/ patterns/ screens/ about.svg templates/ tutorials/ This uses a flat hierarchy for the sake of simplicity. In the CVS codebase, all of these would be placed in inkscape/share/. This way, the installation script could be set up to recursively copy everything in that directory to the target share directory, and we won't need to specify each of the subdirs in that script, which will make maintenance less hassle. For clipart, I think we definitely want to consider being able to tie in with "external" clipart collections. One way would be to simply put symlinks in the clipart dir to those collections. This way, if the flags package were to install into /usr/share/flags-svg/, we'd just symlink there. Possibly similarly for extensions. I think an argument could be made that the tutorials ought to be placed somewhere under the /usr/doc/ or /usr/local/doc/ path, so I'm not 100% certain of having the tutorials located here.

3 4

openoffice fileformat
by Johan Ceuppens 13 Feb '04

13 Feb '04

Hi, I've tested libinkjar and it can read openoffice fileformats now. I'd like to know if I can hook this up and move it to cvs HEAD. I was thinking of bringing up a GUI which greps/seeks the jar file for filenames and letting the user choose which SVGs to open. If you'd like different or additional behaviour, please share with the list. Johan

3 2

Formatting code
by Bob Jamison 13 Feb '04

13 Feb '04

Re: our conversation on Jabber about formatting data for the repr tree, here is some code I appropriated from Apache's APR lib, which can give us total control over formatting output strings. With Mental and Bryce's work in this area, these files might not be necessary, but then again they might be useful, so here they are. One very nice function for the repr tree would be: sp_repr_printf(repr, key, "format", args...) Would relieve the need for a lot of temporary buffers all over the place. Bob /*######################################################################## ## $Id$ ########################################################################*/ /* * Print formatter. This file's purpose is to * allow Inkscape developers to have complete * control over formatted strings. This can be * tweaked to suit. * * This file was basically copied from Apache's * apr_snprintf.c, but with extensive modifications * to ease embedding in a GLib project. Their license * is below. * * Copyright (C) 1999-2003 Authors * * Released under GNU GPL, read the file 'COPYING' for more information * */ /* ==================================================================== * The Apache Software License, Version 1.1 * * Copyright (c) 2000-2003 The Apache Software Foundation. All rights * reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in * the documentation and/or other materials provided with the * distribution. * * 3. The end-user documentation included with the redistribution, * if any, must include the following acknowledgment: * "This product includes software developed by the * Apache Software Foundation (http://www.apache.org/)." * Alternately, this acknowledgment may appear in the software itself, * if and wherever such third-party acknowledgments normally appear. * * 4. The names "Apache" and "Apache Software Foundation" must * not be used to endorse or promote products derived from this * software without prior written permission. For written * permission, please contact apache@...199... * * 5. Products derived from this software may not be called "Apache", * nor may "Apache" appear in their name, without prior written * permission of the Apache Software Foundation. * * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE * DISCLAIMED. IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. * ==================================================================== * * This software consists of voluntary contributions made by many * individuals on behalf of the Apache Software Foundation. For more * information on the Apache Software Foundation, please see * <http://www.apache.org/>. * * The ink_vsnprintf/ink_snprintf functions are based on, and used with the * permission of, the SIO stdio-replacement strx_* functions by Panos * Tsirigotis <panos@...200...> for xinetd. */ #include <math.h> #include <ctype.h> #include <string.h> #include <glib.h> #include "ink-snprintf.h" /*#################################### # D E F I N I T I O N S ####################################*/ /** * Structure used by the variable-formatter routines. */ struct ink_vformatter_buff_t { gchar *curpos;/** The current position */ gchar *endpos;/** The end position of the format string */ }; typedef struct ink_vformatter_buff_t ink_vformatter_buff_t; typedef enum { NO = 0, YES = 1 } boolean_e; #ifndef FALSE #define FALSE 0 #endif #ifndef TRUE #define TRUE 1 #endif #define NUL '\0' #define S_NULL "(null)" #define S_NULL_LEN 6 #define FLOAT_DIGITS 6 #define EXPONENT_LENGTH 10 /* * NUM_BUF_SIZE is the size of the buffer used for arithmetic conversions * * NOTICE: this is a magic number; do not decrease it */ #define NUM_BUF_SIZE 512 /* * cvt.c - IEEE floating point formatting routines for FreeBSD * from GNU libc-4.6.27. Modified to be thread safe. */ /* * ink_ecvt converts to decimal * the number of digits is specified by ndigit * decpt is set to the position of the decimal point * sign is set to 0 for positive, 1 for negative */ #define NDIG 80 /*#################################### # M A C R O S ####################################*/ /* * The INS_CHAR macro inserts a character in the buffer and writes * the buffer back to disk if necessary * It uses the char pointers sp and bep: * sp points to the next available character in the buffer * bep points to the end-of-buffer+1 * While using this macro, note that the nextb pointer is NOT updated. * * NOTE: Evaluation of the c argument should not have any side-effects */ #define INS_CHAR(c, sp, bep, cc) \ { \ if (sp) { \ if (sp >= bep) { \ vbuff->curpos = sp; \ if (flush_func(vbuff)) \ return -1; \ sp = vbuff->curpos; \ bep = vbuff->endpos; \ } \ *sp++ = (c); \ } \ cc++; \ } #define NUM(c) (c - '0') #define STR_TO_DEC(str, num) \ num = NUM(*str++); \ while (g_ascii_isdigit(*str)) \ { \ num *= 10 ; \ num += NUM(*str++); \ } /* * This macro does zero padding so that the precision * requirement is satisfied. The padding is done by * adding '0's to the left of the string that is going * to be printed. We don't allow precision to be large * enough that we continue past the start of s. * * NOTE: this makes use of the magic info that s is * always based on num_buf with a size of NUM_BUF_SIZE. */ #define FIX_PRECISION(adjust, precision, s, s_len) \ if (adjust) { \ gint p = precision < NUM_BUF_SIZE - 1 ? \ precision : NUM_BUF_SIZE - 1; \ while (s_len < p) { \ *--s = '0'; \ s_len++; \ } \ } /* * Macro that does padding. The padding is done by printing * the character ch. */ #define PAD(width, len, ch) \ do \ { \ INS_CHAR(ch, sp, bep, cc); \ width--; \ } \ while (width > len) /* * Prefix the character ch to the string str * Increase length * Set the has_prefix flag */ #define PREFIX(str, length, ch) \ *--str = ch; \ length++; \ has_prefix=YES; /*#################################### # CVT - float-to-ascii conversion ####################################*/ /* buf must have at least NDIG bytes */ static gchar * ink_cvt(gdouble arg, gint ndigits, gint *decpt, gint *sign, gint eflag, gchar *buf) { gint r2; gdouble fi, fj; gchar *p, *p1; if (ndigits >= NDIG - 1) ndigits = NDIG - 2; r2 = 0; *sign = 0; p = &buf[0]; if (arg < 0) { *sign = 1; arg = -arg; } arg = modf(arg, &fi); p1 = &buf[NDIG]; /* * Do integer part */ if (fi != 0) { p1 = &buf[NDIG]; while (p1 > &buf[0] && fi != 0) { fj = modf(fi / 10, &fi); *--p1 = (int) ((fj + .03) * 10) + '0'; r2++; } while (p1 < &buf[NDIG]) *p++ = *p1++; } else if (arg > 0) { while ((fj = arg * 10) < 1) { arg = fj; r2--; } } p1 = &buf[ndigits]; if (eflag == 0) p1 += r2; *decpt = r2; if (p1 < &buf[0]) { buf[0] = '\0'; return (buf); } while (p <= p1 && p < &buf[NDIG]) { arg *= 10; arg = modf(arg, &fj); *p++ = (int) fj + '0'; } if (p1 >= &buf[NDIG]) { buf[NDIG - 1] = '\0'; return (buf); } p = p1; *p1 += 5; while (*p1 > '9') { *p1 = '0'; if (p1 > buf) ++ * --p1; else { *p1 = '1'; (*decpt)++; if (eflag == 0) { if (p > buf) *p = '0'; p++; } } } *p = '\0'; return (buf); } static gchar * ink_ecvt(gdouble arg, gint ndigits, gint *decpt, gint *sign, gchar *buf) { return (ink_cvt(arg, ndigits, decpt, sign, 1, buf)); } static gchar * ink_fcvt(gdouble arg, gint ndigits, gint *decpt, gint *sign, gchar *buf) { return (ink_cvt(arg, ndigits, decpt, sign, 0, buf)); } /* * ink_gcvt - Floating output conversion to * minimal length string */ static gchar * ink_gcvt(gdouble number, gint ndigit, gchar *buf, boolean_e altform) { gint sign, decpt; gchar *p1, *p2; gint i; gchar buf1[NDIG]; p1 = ink_ecvt(number, ndigit, &decpt, &sign, buf1); p2 = buf; if (sign) *p2++ = '-'; for (i = ndigit - 1; i > 0 && p1[i] == '0'; i--) ndigit--; if ((decpt >= 0 && decpt - ndigit > 4) || (decpt < 0 && decpt < -3)) { /* use E-style */ decpt--; *p2++ = *p1++; *p2++ = '.'; for (i = 1; i < ndigit; i++) *p2++ = *p1++; *p2++ = 'e'; if (decpt < 0) { decpt = -decpt; *p2++ = '-'; } else *p2++ = '+'; if (decpt / 100 > 0) *p2++ = decpt / 100 + '0'; if (decpt / 10 > 0) *p2++ = (decpt % 100) / 10 + '0'; *p2++ = decpt % 10 + '0'; } else { if (decpt <= 0) { if (*p1 != '0') *p2++ = '.'; while (decpt < 0) { decpt++; *p2++ = '0'; } } for (i = 1; i <= ndigit; i++) { *p2++ = *p1++; if (i == decpt) *p2++ = '.'; } if (ndigit < decpt) { while (ndigit++ < decpt) *p2++ = '0'; *p2++ = '.'; } } if (p2[-1] == '.' && !altform) p2--; *p2 = '\0'; return (buf); } /* * Convert num to its decimal format. * Return value: * - a pointer to a string containing the number (no sign) * - len contains the length of the string * - is_negative is set to TRUE or FALSE depending on the sign * of the number (always set to FALSE if is_unsigned is TRUE) * * The caller provides a buffer for the string: that is the buf_end argument * which is a pointer to the END of the buffer + 1 (i.e. if the buffer * is declared as buf[ 100 ], buf_end should be &buf[ 100 ]) * * Note: we have 2 versions. One is used when we need to use quads * (conv_10_quad), the other when we don't (conv_10). We're assuming the * latter is faster. */ static gchar * conv_10(gint32 num, gboolean is_unsigned, gboolean *is_negative, gchar *buf_end, gint *len) { gchar *p = buf_end; guint32 magnitude; if (is_unsigned) { magnitude = (guint32) num; *is_negative = FALSE; } else { *is_negative = (num < 0); /* * On a 2's complement machine, negating the most negative integer * results in a number that cannot be represented as a signed integer. * Here is what we do to obtain the number's magnitude: * a. add 1 to the number * b. negate it (becomes positive) * c. convert it to unsigned * d. add 1 */ if (*is_negative) { gint32 t = num + 1; magnitude = ((guint32) -t) + 1; } else magnitude = (guint32) num; } /* * We use a do-while loop so that we write at least 1 digit */ do { guint32 new_magnitude = magnitude / 10; *--p = (gchar) (magnitude - new_magnitude * 10 + '0'); magnitude = new_magnitude; } while (magnitude); *len = buf_end - p; return (p); } static gchar * conv_10_quad(gint64 num, gboolean is_unsigned, gboolean *is_negative, gchar *buf_end, gint *len) { gchar *p = buf_end; guint64 magnitude; /* * We see if we can use the faster non-quad version by checking the * number against the largest long value it can be. If <=, we * punt to the quicker version. */ if ( (num <= ULONG_MAX && is_unsigned) || (num <= LONG_MAX && !is_unsigned) ) return(conv_10( (gint32)num, is_unsigned, is_negative, buf_end, len)); if (is_unsigned) { magnitude = (guint64) num; *is_negative = FALSE; } else { *is_negative = (num < 0); /* * On a 2's complement machine, negating the most negative integer * results in a number that cannot be represented as a signed integer. * Here is what we do to obtain the number's magnitude: * a. add 1 to the number * b. negate it (becomes positive) * c. convert it to unsigned * d. add 1 */ if (*is_negative) { gint64 t = num + 1; magnitude = ((guint64) -t) + 1; } else magnitude = (guint64) num; } /* * We use a do-while loop so that we write at least 1 digit */ do { guint64 new_magnitude = magnitude / 10; *--p = (gchar) (magnitude - new_magnitude * 10 + '0'); magnitude = new_magnitude; } while (magnitude); *len = buf_end - p; return (p); } /* * Convert a floating point number to a string formats 'f', 'e' or 'E'. * The result is placed in buf, and len denotes the length of the string * The sign is returned in the is_negative argument (and is not placed * in buf). */ static gchar * conv_fp(gchar format, gdouble num, boolean_e add_dp, gint precision, gboolean *is_negative, gchar *buf, gint *len) { gchar *s = buf; gchar *p; gint decimal_point; gchar buf1[NDIG]; if (format == 'f') p = ink_fcvt(num, precision, &decimal_point, is_negative, buf1); else /* either e or E format */ p = ink_ecvt(num, precision + 1, &decimal_point, is_negative, buf1); /* * Check for Infinity and NaN */ if (g_ascii_isalpha(*p)) { *len = strlen(p); memcpy(buf, p, *len + 1); *is_negative = FALSE; return (buf); } if (format == 'f') { if (decimal_point <= 0) { *s++ = '0'; if (precision > 0) { *s++ = '.'; while (decimal_point++ < 0) *s++ = '0'; } else if (add_dp) *s++ = '.'; } else { while (decimal_point-- > 0) *s++ = *p++; if (precision > 0 || add_dp) *s++ = '.'; } } else { *s++ = *p++; if (precision > 0 || add_dp) *s++ = '.'; } /* * copy the rest of p, the NUL is NOT copied */ while (*p) *s++ = *p++; if (format != 'f') { gchar temp[EXPONENT_LENGTH]; /* for exponent conversion */ gint t_len; gboolean exponent_is_negative; *s++ = format; /* either e or E */ decimal_point--; if (decimal_point != 0) { p = conv_10((gint32) decimal_point, FALSE, &exponent_is_negative, &temp[EXPONENT_LENGTH], &t_len); *s++ = exponent_is_negative ? '-' : '+'; /* * Make sure the exponent has at least 2 digits */ if (t_len == 1) *s++ = '0'; while (t_len--) *s++ = *p++; } else { *s++ = '+'; *s++ = '0'; *s++ = '0'; } } *len = s - buf; return (buf); } /* * Convert num to a base X number where X is a power of 2. nbits determines X. * For example, if nbits is 3, we do base 8 conversion * Return value: * a pointer to a string containing the number * * The caller provides a buffer for the string: that is the buf_end argument * which is a pointer to the END of the buffer + 1 (i.e. if the buffer * is declared as buf[ 100 ], buf_end should be &buf[ 100 ]) * * As with conv_10, we have a faster version which is used when * the number isn't quad size. */ static gchar * conv_p2(guint32 num, gint nbits, gchar format, gchar *buf_end, gint *len) { gint mask = (1 << nbits) - 1; gchar *p = buf_end; const gchar low_digits[] = "0123456789abcdef"; const gchar upper_digits[] = "0123456789ABCDEF"; const gchar *digits = (format == 'X') ? upper_digits : low_digits; do { *--p = digits[num & mask]; num >>= nbits; } while (num); *len = buf_end - p; return (p); } static gchar * conv_p2_quad(guint64 num, gint nbits, gchar format, gchar *buf_end, gint *len) { gint mask = (1 << nbits) - 1; gchar *p = buf_end; const gchar low_digits[] = "0123456789abcdef"; const gchar upper_digits[] = "0123456789ABCDEF"; const gchar *digits = (format == 'X') ? upper_digits : low_digits; if (num <= ULONG_MAX) return(conv_p2((guint32)num, nbits, format, buf_end, len)); do { *--p = digits[num & mask]; num >>= nbits; } while (num); *len = buf_end - p; return (p); } /*#################################### # F O R M A T T I N G ####################################*/ /* * Do format conversion placing the output in buffer */ static gint ink_vformatter(gint (*flush_func)(ink_vformatter_buff_t *), ink_vformatter_buff_t *vbuff, const gchar *fmt, va_list ap) { gchar *sp; gchar *bep; gint cc = 0; gint i; gchar *s = NULL; gchar *q; gint s_len; gint min_width = 0; gint precision = 0; enum { LEFT, RIGHT } adjust; gchar pad_char; gchar prefix_char; gdouble fp_num; gint64 i_quad = (gint64) 0; guint64 ui_quad; gint32 i_num = (gint32) 0; guint32 ui_num; gchar num_buf[NUM_BUF_SIZE]; gchar char_buf[2]; /* for printing %% and %<unknown> */ enum var_type_enum { IS_QUAD, IS_LONG, IS_SHORT, IS_INT }; enum var_type_enum var_type = IS_INT; /* * Flag variables */ boolean_e alternate_form; boolean_e print_sign; boolean_e print_blank; boolean_e adjust_precision; boolean_e adjust_width; gboolean is_negative; sp = vbuff->curpos; bep = vbuff->endpos; while (*fmt) { if (*fmt != '%') { INS_CHAR(*fmt, sp, bep, cc); } else { /* * Default variable settings */ boolean_e print_something = YES; adjust = RIGHT; alternate_form = print_sign = print_blank = NO; pad_char = ' '; prefix_char = NUL; fmt++; /* * Try to avoid checking for flags, width or precision */ if (!g_ascii_islower(*fmt)) { /* * Recognize flags: -, #, BLANK, + */ for (;; fmt++) { if (*fmt == '-') adjust = LEFT; else if (*fmt == '+') print_sign = YES; else if (*fmt == '#') alternate_form = YES; else if (*fmt == ' ') print_blank = YES; else if (*fmt == '0') pad_char = '0'; else break; } /* * Check if a width was specified */ if (g_ascii_isdigit(*fmt)) { STR_TO_DEC(fmt, min_width); adjust_width = YES; } else if (*fmt == '*') { min_width = va_arg(ap, int); fmt++; adjust_width = YES; if (min_width < 0) { adjust = LEFT; min_width = -min_width; } } else adjust_width = NO; /* * Check if a precision was specified */ if (*fmt == '.') { adjust_precision = YES; fmt++; if ( g_ascii_isdigit (*fmt)) { STR_TO_DEC(fmt, precision); } else if (*fmt == '*') { precision = va_arg(ap, int); fmt++; if (precision < 0) precision = 0; } else precision = 0; } else adjust_precision = NO; } else adjust_precision = adjust_width = NO; /* * Modifier check */ if (*fmt == 'q') { var_type = IS_QUAD; fmt++; } else if (*fmt == 'l') { var_type = IS_LONG; fmt++; } else if (*fmt == 'h') { var_type = IS_SHORT; fmt++; } else { var_type = IS_INT; } /* * Argument extraction and printing. * First we determine the argument type. * Then, we convert the argument to a string. * On exit from the switch, s points to the string that * must be printed, s_len has the length of the string * The precision requirements, if any, are reflected in s_len. * * NOTE: pad_char may be set to '0' because of the 0 flag. * It is reset to ' ' by non-numeric formats */ switch (*fmt) { case 'u': if (var_type == IS_QUAD) { i_quad = va_arg(ap, guint64); s = conv_10_quad(i_quad, 1, &is_negative, &num_buf[NUM_BUF_SIZE], &s_len); } else { if (var_type == IS_LONG) i_num = (gint32) va_arg(ap, guint32); else if (var_type == IS_SHORT) i_num = (gint32) (unsigned short) va_arg(ap, unsigned int); else i_num = (gint32) va_arg(ap, unsigned int); s = conv_10(i_num, 1, &is_negative, &num_buf[NUM_BUF_SIZE], &s_len); } FIX_PRECISION(adjust_precision, precision, s, s_len); break; case 'd': case 'i': if (var_type == IS_QUAD) { i_quad = va_arg(ap, gint64); s = conv_10_quad(i_quad, 0, &is_negative, &num_buf[NUM_BUF_SIZE], &s_len); } else { if (var_type == IS_LONG) i_num = (gint32) va_arg(ap, gint32); else if (var_type == IS_SHORT) i_num = (gint32) (short) va_arg(ap, int); else i_num = (gint32) va_arg(ap, int); s = conv_10(i_num, 0, &is_negative, &num_buf[NUM_BUF_SIZE], &s_len); } FIX_PRECISION(adjust_precision, precision, s, s_len); if (is_negative) prefix_char = '-'; else if (print_sign) prefix_char = '+'; else if (print_blank) prefix_char = ' '; break; case 'o': if (var_type == IS_QUAD) { ui_quad = va_arg(ap, guint64); s = conv_p2_quad(ui_quad, 3, *fmt, &num_buf[NUM_BUF_SIZE], &s_len); } else { if (var_type == IS_LONG) ui_num = (guint32) va_arg(ap, guint32); else if (var_type == IS_SHORT) ui_num = (guint32) (unsigned short) va_arg(ap, unsigned int); else ui_num = (guint32) va_arg(ap, unsigned int); s = conv_p2(ui_num, 3, *fmt, &num_buf[NUM_BUF_SIZE], &s_len); } FIX_PRECISION(adjust_precision, precision, s, s_len); if (alternate_form && *s != '0') { *--s = '0'; s_len++; } break; case 'x': case 'X': if (var_type == IS_QUAD) { ui_quad = va_arg(ap, guint64); s = conv_p2_quad(ui_quad, 4, *fmt, &num_buf[NUM_BUF_SIZE], &s_len); } else { if (var_type == IS_LONG) ui_num = (guint32) va_arg(ap, guint32); else if (var_type == IS_SHORT) ui_num = (guint32) (unsigned short) va_arg(ap, unsigned int); else ui_num = (guint32) va_arg(ap, unsigned int); s = conv_p2(ui_num, 4, *fmt, &num_buf[NUM_BUF_SIZE], &s_len); } FIX_PRECISION(adjust_precision, precision, s, s_len); if (alternate_form && i_num != 0) { *--s = *fmt; /* 'x' or 'X' */ *--s = '0'; s_len += 2; } break; case 's': s = va_arg(ap, gchar *); if (s != NULL) { if (!adjust_precision) { s_len = strlen(s); } else { /* From the C library standard in section 7.9.6.1: * ...if the precision is specified, no more then * that many characters are written. If the * precision is not specified or is greater * than the size of the array, the array shall * contain a null character. * * My reading is is precision is specified and * is less then or equal to the size of the * array, no null character is required. So * we can't do a strlen. * * This figures out the length of the string * up to the precision. Once it's long enough * for the specified precision, we don't care * anymore. * * NOTE: you must do the length comparison * before the check for the null character. * Otherwise, you'll check one beyond the * last valid character. */ const gchar *walk; for (walk = s, s_len = 0; (s_len < precision) && (*walk != '\0'); ++walk, ++s_len); } } else { s = S_NULL; s_len = S_NULL_LEN; } pad_char = ' '; break; case 'f': case 'e': case 'E': fp_num = va_arg(ap, double); /* * We use &num_buf[ 1 ], so that we have room for the sign */ s = NULL; #ifdef HAVE_ISNAN if (isnan(fp_num)) { s = "nan"; s_len = 3; } #endif #ifdef HAVE_ISINF if (!s && isinf(fp_num)) { s = "inf"; s_len = 3; } #endif if (!s) { s = conv_fp(*fmt, fp_num, alternate_form, (adjust_precision == NO) ? FLOAT_DIGITS : precision, &is_negative, &num_buf[1], &s_len); if (is_negative) prefix_char = '-'; else if (print_sign) prefix_char = '+'; else if (print_blank) prefix_char = ' '; } break; case 'g': case 'G': if (adjust_precision == NO) precision = FLOAT_DIGITS; else if (precision == 0) precision = 1; /* * * We use &num_buf[ 1 ], so that we have room for the sign */ s = ink_gcvt(va_arg(ap, double), precision, &num_buf[1], alternate_form); if (*s == '-') prefix_char = *s++; else if (print_sign) prefix_char = '+'; else if (print_blank) prefix_char = ' '; s_len = strlen(s); if (alternate_form && (q = strchr(s, '.')) == NULL) { s[s_len++] = '.'; s[s_len] = '\0'; /* delimit for following strchr() */ } if (*fmt == 'G' && (q = strchr(s, 'e')) != NULL) *q = 'E'; break; case 'c': char_buf[0] = (char) (va_arg(ap, int)); s = &char_buf[0]; s_len = 1; pad_char = ' '; break; case '%': char_buf[0] = '%'; s = &char_buf[0]; s_len = 1; pad_char = ' '; break; case 'n': if (var_type == IS_QUAD) *(va_arg(ap, gint64 *)) = cc; else if (var_type == IS_LONG) *(va_arg(ap, long *)) = cc; else if (var_type == IS_SHORT) *(va_arg(ap, short *)) = cc; else *(va_arg(ap, int *)) = cc; print_something = NO; break; /* * This is where we extend the printf format, with a second * type specifier */ case 'p': switch(*++fmt) { /* * If the pointer size is equal to or smaller than the size * of the largest unsigned int, we convert the pointer to a * hex number, otherwise we print "%p" to indicate that we * don't handle "%p". */ case 'p': #ifdef VOID_P_IS_QUAD if (sizeof(void *) <= sizeof(guint64)) { ui_quad = (guint64) va_arg(ap, void *); s = conv_p2_quad(ui_quad, 4, 'x', &num_buf[NUM_BUF_SIZE], &s_len); } #else if (sizeof(void *) <= sizeof(guint32)) { ui_num = (guint32) va_arg(ap, void *); s = conv_p2(ui_num, 4, 'x', &num_buf[NUM_BUF_SIZE], &s_len); } #endif else { s = "%p"; s_len = 2; prefix_char = NUL; } pad_char = ' '; break; case 'T': char_buf[0] = '0'; s = &char_buf[0]; s_len = 1; pad_char = ' '; break; case NUL: /* if %p ends the string, oh well ignore it */ continue; default: s = "bogus %p"; s_len = 8; prefix_char = NUL; (void)va_arg(ap, void *); /* skip the bogus argument on the stack */ break; } break; case NUL: /* * The last character of the format string was %. * We ignore it. */ continue; /* * The default case is for unrecognized %'s. * We print %<char> to help the user identify what * option is not understood. * This is also useful in case the user wants to pass * the output of format_converter to another function * that understands some other %<char> (like syslog). * Note that we can't point s inside fmt because the * unknown <char> could be preceded by width etc. */ default: char_buf[0] = '%'; char_buf[1] = *fmt; s = char_buf; s_len = 2; pad_char = ' '; break; } if (prefix_char != NUL && s != S_NULL && s != char_buf) { *--s = prefix_char; s_len++; } if (adjust_width && adjust == RIGHT && min_width > s_len) { if (pad_char == '0' && prefix_char != NUL) { INS_CHAR(*s, sp, bep, cc); s++; s_len--; min_width--; } PAD(min_width, s_len, pad_char); } /* * Print the string s. */ if (print_something == YES) { for (i = s_len; i != 0; i--) { INS_CHAR(*s, sp, bep, cc); s++; } } if (adjust_width && adjust == LEFT && min_width > s_len) PAD(min_width, s_len, pad_char); } fmt++; } vbuff->curpos = sp; return cc; } static int snprintf_flush(ink_vformatter_buff_t *vbuff) { /* if the buffer fills we have to abort immediately, there is no way * to "flush" an ink_snprintf... there's nowhere to flush it to. */ return -1; } gint ink_snprintf(gchar *buf, gint len, const gchar *format, ...) { gint cc; va_list ap; ink_vformatter_buff_t vbuff; if (len == 0) { /* NOTE: This is a special case; we just want to return the number * of chars that would be written (minus \0) if the buffer * size was infinite. We leverage the fact that INS_CHAR * just does actual inserts iff the buffer pointer is non-NULL. * In this case, we don't care what buf is; it can be NULL, since * we don't touch it at all. */ vbuff.curpos = NULL; vbuff.endpos = NULL; } else { /* save one byte for nul terminator */ vbuff.curpos = buf; vbuff.endpos = buf + len - 1; } va_start(ap, format); cc = ink_vformatter(snprintf_flush, &vbuff, format, ap); va_end(ap); if (len != 0) { *vbuff.curpos = '\0'; } return (cc == -1) ? (gint)len : cc; } gint ink_vsnprintf(gchar *buf, gint len, const gchar *format, va_list ap) { gint cc; ink_vformatter_buff_t vbuff; if (len == 0) { /* See above note */ vbuff.curpos = NULL; vbuff.endpos = NULL; } else { /* save one byte for nul terminator */ vbuff.curpos = buf; vbuff.endpos = buf + len - 1; } cc = ink_vformatter(snprintf_flush, &vbuff, format, ap); if (len != 0) { *vbuff.curpos = '\0'; } return (cc == -1) ? (gint)len : cc; } #ifndef __INK_SNPRINTF_H__ #define __INK_SNPRINTF_H__ /*######################################################################## ## $Id$ ########################################################################*/ /* * Print formatter. This file's purpose is to * allow Inkscape developers to have complete * control over formatted strings. This can be * tweaked to suit. * * Copyright (C) 1999-2003 Authors * * Released under GNU GPL, read the file 'COPYING' for more information */ #include <glib.h> #include <stdarg.h> /** * Create a formatted string with a given format string * and a standard variable argument list. */ gint ink_snprintf(gchar *buf, gint len, const gchar *format, ...); /** * Create a formatted string with a given format string * and argument list. */ gint ink_vsnprintf(gchar *buf, gint len, const gchar *format, va_list ap); #endif /* __INK_SNPRINTF_H__ */

1 0

Re: [Inkscape-devel] Some fairly random feature reqs + bug reports.
by bulia byak 13 Feb '04

13 Feb '04

>>>What happened in the Fill/stroke dialog? the entry >>>boxs for RGB values on the wheel tab are gone >> >> >>That's temporary, we plan to restore them, although not RGB - the wheel is >>HSV and it will have HSV fields, if you need RGB switch to the RGB >>sliders. > >Well, that I'm not sure of. Jon, I do not mean restore the old system picker, but to add HSV fields to your new picker, right above the alpha spinbutton. _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

RE: [Inkscape-devel] Some fairly random feature reqs + bug reports.
by bulia byak 13 Feb '04

13 Feb '04

Thanks for the feedback! > First, if I've got a new session of inkscape, and i >do file>open, why does it open it into a new window >rather than the blank one I've got in front of me? >every time it does it it strikes me as odd. Yes, it's in the plans. Maybe next version. > Second up, have you considered making use of the >boolean ops to add paintbrush/eraser tools ala flash? >the functionality is all there just spread across >several tools currently. The way I've been simulating >it currently is to use the calligraphy tool to 'paint' >an area, then whenever you paint additional >areas immediately switch to select and ctrl++ it to >the last section. Not sure about usefulness of this. The vector approach is to keep things separate and at a high level of abstraction for as long as possible, and only merge/flatten/convert to paths as the last stage if possible. So if I need this functionality, I would actually prefer to have shapes separate to I can move and adjust them, and do a union or difference on them if necessary (but I think in many cases, just grouping them together would be convenient enough). On the other hand, what you propose may be simple enough to implement, but we need to do a paint tool with user-definable strokes first, otherwise the booleans are of little use IMHO. > Another feature that I'd love to see would be >variable offsets, ie a dynamic offset with >node weighting allowing the offset to be modified at >each node so I could generate a line of varying >thickness (like the calligraphy tool) but still >editable by the basic centreline path. This would be a >godsend for doing comic style drawings where the line >weight is important. I know that ones a big task, just >putting the idea out there. Yes, useful, but perhaps more uesful in the short term and easier to implement is a variable-width stroke with some predefined shapes. I think our developers had some prototypes of this. > With regards to the 0.37 release, the builds i >downloaded from Bobs site (which is really useful, >both for builds and build instructions, cheers Bob) >seemed to have problems with the keyboard shortcuts >being messed up, the dialog ones were all muddled, >(for instance Ctrl+Shift+M was opening fill/stroke) >just checked the latest one there and its still the >case.(zip file Inkscape0402120319.zip) Very strange. Anyone else experiencing this? >What happened in the Fill/stroke dialog? the entry >boxs for RGB values on the wheel tab are gone That's temporary, we plan to restore them, although not RGB - the wheel is HSV and it will have HSV fields, if you need RGB switch to the RGB sliders. >as is >the dropper icon, both of which were really useful. Dropper on the dialog is redundant, you can use the dropper tool >(its a pain having to switch tool to do a dropper >action, What's the problem with the dropper? We discussed moving the dropper to the dialog and removing the tool, but then decided that the dropper will be very useful without the dialog, when we have a current color indicator and a palette. >btw, anyone got any suggestions as to the best place >to get started with getting familiar with the code? Read developer's howtos on Wiki - there's quite a lot of info there, although it's unorganized. Then look into the code and try to find what seems easiest - I don't know what it may be for you :) I started with adding new keyboard shortcuts, that was easy and fun, and gradually learned many parts of the codebase from there. _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

3 2

RE: [Inkscape-devel] openoffice fileformat
by bulia byak 13 Feb '04

13 Feb '04

Johan, >I've tested libinkjar and it can read openoffice fileformats now. >I'd like to know if I can hook this up and move it to cvs HEAD. Could you please add a paragraph describing what you did to http://inkscape.org/cgi-bin/wiki.pl?ReleaseNotes _________________________________________________________________ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

1 0

RE: [Inkscape-devel] Some fairly random feature reqs + bug reports.
by bulia byak 12 Feb '04

12 Feb '04

>fair enough, however its a tool that flash users have >got used to, and is a good way of generating and >manipulating complex shapes in vector form without the >user needing to worry about bezier handles, adding >nodes etc. If you can use paint you can use that tool. You're right, this might be useful, after we have a "hose" tool that will draw by spitting out shapes. Could you please add a RFE to the tracker? >see >http://members.lycos.co.uk/simarilius/weirdness.jpg I'm stunned. How can this happen - anyone having any ideas? Mental, what do you think? > > HSV and it will have HSV fields, if you need RGB > > switch to the RGB sliders. > > > >they dont allow for 0-255 values like most art >programs tho, which is what was nice about the wheel >tab, it had commonality with a lot of other art >programs. I think you're right, we should switch all sliders on all color picker tabs from 0.00..1.00 to 0..255 (integer). It's more useful and gives more precision anyway. However, we must leave the possibility to use the old format as well, by typing e.g. "50%" or "0.5" which, after you press Enter, becomes 127. Again, could you please add an RFE? (Or if you wanted to try contributing, here's a nice little project for you to start :) >selecting the dropper tool switches you out of the >current context, hence you have to swap back. The one >in the wheel panel doesnt, it drops straight back to >whatever you were doing before, which is easier, and >less disruptive to the workflow. Anyway, i go into the >fill dialog to pick a color, isnt it logical to be >able to access the tool to do that from in there? Perhaps you are right, and we'll need to rethink the dropper role and position. I'd say let's do the more urgent color stuff first, which in my opition is a palette and a current color indicator, and then the perfect place of the dropper will be easier to find. _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

1 0

Re: [Inkscape-devel] Strange bug in Windows build?
by bulia byak 12 Feb '04

12 Feb '04

>). But you'll probably see it in the attached svg file that there are >freakish numbers showing... This used to be a small default values star... >:( Thanks, that helps. It would be even better if you could send a couple of "before" and "after" files that are only different by the offending convert to path operation, so we could see exactly what points were displaced and what were not. Please use a simplest possible shape for this (ellipse or square, not star). Being unable to reproduce, I cannot guarantee fixing this, but that would still be helpful. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 4

About
by Daniel Díaz 12 Feb '04

12 Feb '04

I forgot to mention it again: The About Inkscape window is still showing "Frederico Mena", but it should read "Federico Mena-Quintero" (note only one R in Federico). Oh, by the way, when I right click in text written in the canvas, why should I care more about the Item Properties than Text and Font? I think it should be included in there. Greetings! Daniel Díaz yosoy@...31...

1 0

Fedora-1 rpms on sourceforge
by Peter Linnell 12 Feb '04

12 Feb '04

Hi all, Just a quick note to let yet know the Fedora-1 rpms are in the queue for loading on sourceforge. Peter aka mrdocs

2 1

Some fairly random feature reqs + bug reports.
by John Cliff 12 Feb '04

12 Feb '04

Hello all, been playing with both sodipodi and inkscape for a few weeks and have finally got round to looking at the source code. Have a couple of suggestions for features / usability issues. First, if I've got a new session of inkscape, and i do file>open, why does it open it into a new window rather than the blank one I've got in front of me? every time it does it it strikes me as odd. Second up, have you considered making use of the boolean ops to add paintbrush/eraser tools ala flash? the functionality is all there just spread across several tools currently. The way I've been simulating it currently is to use the calligraphy tool to 'paint' an area, then whenever you paint additional areas immediately switch to select and ctrl++ it to the last section. To use it as an eraser just paint over the bits you want to delete and do a subtract. What would be good would be if there was a tool that would automatically do the boolean to the current selection. (and possibly a simplify as well.) It could do with having constant stroke width rather than the calligriphic stuff tho, preferably of selectable width from the secondary toolbar, but still with the realtime preview of whats being drawn. I'd love to be able to try this myself, but am struggling even getting the codebase to compile atm, didnt help that the tar was lacking the mingw makefiles. Am a delphi person normally and havent touched C/C++ since uni, so am more than a little rusty. :) Another feature that I'd love to see would be variable offsets, ie a dynamic offset with node weighting allowing the offset to be modified at each node so I could generate a line of varying thickness (like the calligraphy tool) but still editable by the basic centreline path. This would be a godsend for doing comic style drawings where the line weight is important. I know that ones a big task, just putting the idea out there. With regards to the 0.37 release, the builds i downloaded from Bobs site (which is really useful, both for builds and build instructions, cheers Bob) seemed to have problems with the keyboard shortcuts being messed up, the dialog ones were all muddled, (for instance Ctrl+Shift+M was opening fill/stroke) just checked the latest one there and its still the case.(zip file Inkscape0402120319.zip) What happened in the Fill/stroke dialog? the entry boxs for RGB values on the wheel tab are gone as is the dropper icon, both of which were really useful. (its a pain having to switch tool to do a dropper action, and a lot of apps give their colors in RGB, so being able to enter the values from somewhere else was good.) It is a better size now, but that tab still has a fair bit of grey space in it which could be used. Also on the this build the icons are red crosses on the tabs and join&cap selectors in that dialog. Really good work on the tutorial btw, is really helpful, and looks great. On the website the mailing lists page doesnt link to the user list archive, which would be good. sorry to hit you with so much stuff at once, just been looking at it in increasing detail the last few days. keep up the good work John Cliff btw, anyone got any suggestions as to the best place to get started with getting familiar with the code? __________________________________ Do you Yahoo!? Yahoo! Finance: Get your refund fast by filing online. http://taxes.yahoo.com/filing.html

1 0

Inkscape 0.37 RPMS
by Bryce Harrington 12 Feb '04

12 Feb '04

I've uploaded i386 and i686 versions of the Mandrake RPM I mentioned yesterday. They're available from the Inkscape SourceForge download page (go to http://www.inkscape.org/ and click Download). The feedback I've received is that there may be some dependency issues, which I'm working on currently. I would appreciate additional feedback though, as that might help clarify what changes are needed. Bryce

1 0

N/A
by jon 12 Feb '04

12 Feb '04

I propose that we should put our OFFICIAL RELEASE off till SUNDAY NIGHT/MONDAY MORNING, so that we avoid any crossover with SODIPODI. This won't hold up any development, but will help us further differentiate ourselves from SODI's release today. It really doesn't matter who comes out with what first, but really how the program works and how well user's like it. Do you all agree with this delay in publishing the announcement? It won't hold up development at all, but just delay PR. Also, I wonder if we should have a look at the SODIPODI codebase and co-opt any new code that might be valuable. I think competition is very good for both projects, honestly, but is something to wrestle with in terms of release dates. Jon

2 1

Re: [Inkscape-devel] Strange bug in Windows build?
by bulia byak 12 Feb '04

12 Feb '04

I did not notice the new comment in that bug saying that glib 2.3.1 resolves the problem: http://sourceforge.net/tracker/index.php?func=detail&aid=857311&group_id=93… Bob, what it takes to package inkscape-windows with this version of glib? Is this possible? _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

3 2

RE: [Inkscape-devel] N/A
by bulia byak 12 Feb '04

12 Feb '04

>Also, I wonder if we should have a look at the SODIPODI codebase and co-opt >any new code that might be valuable. I was doing it all this time, too bad nobody (except Mental) joined me. Thus, xml:space and page background color support are from SP (with my improvements), as well as many fixes. I urge everyone to do more of that. Just make a diff between SP releases and see what was changed, and if you see a change that you understand and think might be good for us too, port. It's actually a pretty interesting job that might help you get a good understanding of the codebase. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

fr.po (again)
by unknown＠example.com 11 Feb '04

11 Feb '04

Hi, I join an update fr.po . Fred.

1 0

Files in SF
by Ted Gould 11 Feb '04

11 Feb '04

Okay, SF was up for a few minutes, and I seem to have gotten files into it. I've also updated the front page to link there. Here's the link: https://sourceforge.net/project/showfiles.php?group_id=93438&package_id=991… Also, I'm not sure that I did the whole GPG thing right. I just did a 'gpg -s tarball' and it made the .gpg file. But I thought it would have been smaller. Does this seem right? --Ted

2 1

Re: [Inkscape-devel] Strange bug in Windows build?
by bulia byak 11 Feb '04

11 Feb '04

>>Thanks, that helps. It would be even better if you could send a couple of >>"before" and "after" files that are only different by the offending >>convert to path operation, > >Here are the before and after svg's. The original has a 100.0x100.0 square. Thanks - now that I see it, I think this is actually a known bug: http://sourceforge.net/tracker/index.php?func=detail&aid=857311&group_id=93… The reason being that your Windows (Icelandic?) seems to use comma instead of period for fractional numbers (e.g. 927,38287 instead of 927.38287), and this drives Inkscape crazy. We'll need to focus on this bug, among others, for the next bug-fixing release. Please attach your before and after examples with comments to that bug, they may be helpful. As a temporary workaround, you can change the number format in your Windows settings - please let me know if this helps. _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

1 0

Fedora BuildRequires:
by Peter Linnell 11 Feb '04

11 Feb '04

BuildRequires: XFree86-devel >= 0:4.3.0 BuildRequires: libgnomeprintui22-devel BuildRequires: libpng-devel BuildRequires: desktop-file-utils BuildRequires: gettext BuildRequires: libart_lgpl-devel >= 2.3.10 BuildRequires: freetype-devel BuildRequires: libxml2-devel >= 2-2.4.24 BuildRequires: gtk2-devel BuildRequires: pango-devel BuildRequires: atk-devel BuildRequires: pkgconfig ./configure --with-gnome-print **Thanks Johan :)

1 0

0.37 Release Tarball
by Bryce Harrington 11 Feb '04

11 Feb '04

Hi all, Ted has tagged and cut the tarball for the 0.37 release. At this time we're asking for packagers to download and produce the packages according to the instructions on the CreatingDists page in Wiki. http://www.inkscape.org/cgi-bin/wiki.pl?CreatingDists Since SF is down, we've uploaded the tarball to the website temporarily: http://www.inkscape.org/inkscape-0.37.tar.gz Make sure to follow the naming scheme when making your packages, and then upload them to SourceForge's file manager, or post them someplace convenient and point me, nathan, ted, or one of the other SF admins at them to pick up. Bryce

5 7

RE: [Inkscape-devel] Strange bug in Windows build?
by bulia byak 11 Feb '04

11 Feb '04

>The thing is that if I use the freehand tool ( or any other tool than the >predefined shapes ), as soon as I release the mouse button the shape >literally goes haywire, that is, it becomes a big spindly mess. Never seen anything like it. The important questions is, do the actual nodes of the path become displaced to (0, -infinity), or is it only a rendering problem while the nodes are where they need to be? Can you save such a broken file? If so, is it displayed broken when you reopen it in the win build? or in linux build? If the saved SVG really becomes broken, please send it to me (and file a bug attaching your screenshots and the broken SVG). Otherwise this problem may be hard to solve because it will have to do with the renderer which is much more difficult to debug. _________________________________________________________________ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

RE: [Inkscape-devel] Possibility of tutorial translation
by bulia byak 11 Feb '04

11 Feb '04

>May be it is reasonable to add gettext function to sp_help_tutorial: > >sp_help_tutorial (void) >{ > sp_file_open (INKSCAPE_PIXMAPDIR _("/tutorial.svg"), NULL); > >Then it will be possible to support translations of tutorial. Good idea, I'll do that. If you can provide a translation of the tutorial that would be even better :) _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

2 1

accelerated scrolling
by bulia byak 11 Feb '04

11 Feb '04

The tutorial and keys reference look nice but are a pain in the ass to scroll, especially in zoom-in. Until today. * Previously, Inkscape canvas tended to "stick" after mouse release when you were panning a complex document by middle mouse button. This is now fixed. * Accelerated scrolling: if you press and hold Ctrl+arrow to scroll canvas, Inkscape will gradually speed up scrolling. This makes navigating in complex documents much faster, without losing precision. The power of this acceleration is given by the pref value options.scrollingacceleration. I set it to 2 by default; please test and report which value works best for you (edit your ~/.inkscape/preferences.xml). The value of 1 means no acceleration, but even at 1 it still scrolls faster than before because it gobbles sequences of keypresses and, for n key events, will scroll once by scroll_step*n instead of scrolling n times by scroll_step. I plan to apply the same acceleration approach to areas other than scrolling, such as zooming and nudging by keyboard. I will also fix sticking after mouse release in dragging objects and nodes, freehand drawing, etc. Suggestions welcome. _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

1 0

Remaining Tasks for 0.37 Release
by Bryce Harrington 11 Feb '04

11 Feb '04

Here's the specific things to do before the feature freeze: * Review #if 0 code in bonobo/ * Review #if 0 code in desktop*.cpp and svg-view.cpp files [ted] * Review #if 0 code in sp-gradient.cpp, sp-root.cpp [tvon] * Review #if 0 code in file.h, interface.cpp, geom.cpp, inkscape.cpp, and style.cpp [ishmal] * Review #if 0 code in *-chemistry.cpp and seltrans.cpp [pjrm] * Review #if 0 code in svg/ * Review #if 0 code in display/ [mental] * Review #if 0 code in xml/ * Convert all tabs into spaces in codebase * Rename all 'SPFooBar' routines to 'FooBar' and put into namespaces [mental] * Update coding-style.txt from the Coding_Style page in wiki When all of these are done, we will feature-freeze, and the focus will be on bug fixing. There are 10 bugs rated critical, that will need fixing. 5 of these do not yet have an owner so we need volunteers to take care of them. We also need to review the lower priority bugs to see if there are additional critical ones that will need fixing. Once the critical bugs are fixed, we will establish the stable branch for 0.37 and the release team can get to work on packaging and creation of binaries. Once we have binaries for Windows, OSX, and a couple common linux distros, we can begin sending out announcements. :-) Bryce

7 13

Re: [Inkscape-devel] Grid toggle / snapping keyboard shortcut
by bulia byak 10 Feb '04

10 Feb '04

> > Would it be feasible to have the # key toggle both the grid visibility > > and grid snapping at the same time? > >Before implementing, we should consider what the key should do if the >grid is currently enabled and not snapping, hide grid, don't touch snapping >or vice versa. show grid, don't touch snapping imho _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

Grid toggle / snapping keyboard shortcut
by Bryce Harrington 10 Feb '04

10 Feb '04

Bulia, I've been making good use of the # shortcut for turning the grid on and off. Quite handy. However, every time I put the grid on, I also want to snap to the grid, yet afaik there's no shortcut key for it, so I have to open the edit window settings anyway, so the value of the # shortcut key is lost. Would it be feasible to have the # key toggle both the grid visibility and grid snapping at the same time? If it is feasible, what do you think about setting it to work that way? I think it'd be a big time saver for those of us interested in using inkscape for technical drawing. Bryce

4 5

POPT bugfix
by Bob Jamison 10 Feb '04

10 Feb '04

I don't know how, but the file association stuff wasn't working last night. Since I really, really wanted that to work, I looked at main.cpp to see what Win32 needed to be in sync. Main.cpp, as it is, needs WITH_POPT badly, and we haven't been using it on Win32. Until now. I stayed up all night and made a static build of libpopt.a, and added it to the canned Gtk/Gnu win32 libs. The Gnu-Win32 web site's version of popt had unwanted dependencies. Now the command-line parsing on Win32 is the same as for the Unices, so things should be relatively unbreakable now. I also added inkscape.nsi, which is an NSIS (Nullsoft Installer) script which performs some minimalist installer/uninstaller functions. I still need to add the Bitstream Vera fonts to it. Bob

2 2

FINAL RELEASE OF 0.37 HAPPENING NOW -- HOLD CVS COMMITS
by Jonathan Phillips 10 Feb '04

10 Feb '04

Hello ALL, The tree is cut and the 0.37 package process is beginning. Only very very very serious fixes are allowed, and get onto chat b4 any commits. We have closed most of the critical bugs, and any known bugs we pushed into the NEWS and RELEASE NOTES. We will probably issue a maintenance release like 0.37.1 after a few little features are ironed out post our 0.37 release, but we are getting this thing out of the gate tonight. Thanks, Jon -- Jonathan Phillips <jon@...15...>

2 1

Possibility of tutorial translation
by Alexey Remizov 10 Feb '04

10 Feb '04

Hello. May be it is reasonable to add gettext function to sp_help_tutorial: sp_help_tutorial (void) { sp_file_open (INKSCAPE_PIXMAPDIR _("/tutorial.svg"), NULL); Then it will be possible to support translations of tutorial. -- С уважением. | WBR. | Алексей. | Alexey. | ----------------------------| mailto:alexey@...188... | jabber:remizov@...189... | ----------------------------'

1 0

Re: [Inkscape-devel] tutorial, terms
by bulia byak 10 Feb '04

10 Feb '04

>Here are the Bitstream Vera files. I suggest that we make a >/fonts toplevel directory in the CVS tree and add them. They >add up to 564kb. Maybe add them only for Windows? Reasons: 1) Linux users are more likely to have them already, and 2) they are generally less happy when a program installs something except itself. >I haven't looked, but isn't it possible in X to add a >font for a specific application? Just add it to the >font path? If so, it would make installing Inkscape >on X a lot easier On a modern system, you just drop them into ~/.fonts. But I don't know if it's acceptable for an RPM or install script to do this. As for Windows, yes, ideally it should install these fonts, and at the very least, the installer should provide the fonts and tell the user to install them. _________________________________________________________________ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

6 6

Release Candidate + Consensus
by Jonathan Phillips 10 Feb '04

10 Feb '04

Hello All, We got 2 more critical bugs out of the way and that was our target range for the release. Soooo, with developer consensus we can move forward with cutting a release (this is what we decided in chat just now). E-mail the list if problems or get online so we can hammer it out ASAP, for tonight is the night! Jon -- Jonathan Phillips <jon@...15...>

1 0

0.37 Release objectives met
by Bryce Harrington 10 Feb '04

10 Feb '04

The task and bugcount objectives for the 0.37 release have now been met, so we are ready to tag the stable branch and begin the release process this evening. At this time, you can help by running the app and banging on it to see if any of the recent bug fixes have introduced new problems that will need fixing prior to the release. Hopefully there's nothing. We have reduced the critical bug count to 3, and thus met our objective. Two are difficult to do and will take significant time to do properly, and the third is in regards to the nightly builds posted to the website so is completely orthogonal to the application itself. Once the release tarball is posted this evening, we will allocate 2-3 days for folks to create and test packages for this release, and then will put out announcements to as many places as will listen. :-) Bryce

1 0

Status Update for Inkscape release work
by Bryce Harrington 09 Feb '04

09 Feb '04

Here are the remaining code cleanup tasks: ** Review #if 0 code in sp-gradient.cpp, sp-root.cpp [tvon] ** Review #if 0 code in *-chemistry.cpp and seltrans.cpp [pjrm] ** (DONE) Review #if 0 code in svg/ [bryce] ** (DONE) Review #if 0 code in display/ [mental] ** (DONE) Review #if 0 code in helper/ [mental] ** (DONE) Review #if 0 code in xml/ [mental] I've mailed out a detailed post about releasing packages, and updated the CreatingDists Wiki page. We are down to 7 critical bugs that will need to be addressed prior to our final release: 891852 saves to old location after save-as [ted] 850971 (DONE) Crash while deleting namedview in XML editor [bryce] 852976 win32: mouse becoming unresponsive 864791 XML editor: crash when canceling quit [mental] 863373 ./configure not present in nightly tarball snapshot 863371 (DONE) --gnome-print option [jceuppen] 881397 too big objects crash 887428 R&B channels reversed on win32 when printing [ishmal] 871582 path display freeze [mental] I think we'd all love to see 100% of these fixed, but I figure if we can at least get 2/3rds of them addressed, then we should go ahead with a release. So if we can close 4 more, we should be there. 887428 is fixed but needs to be verified on Windows. 71582 is probably going to be more work than we can do for this release so we may want to leave it 'til next release. Mental, is there a quick hack that could be done to at least prevent Inkscape from freezing up when zooming out? Could you write up your thoughts on how the code should be rewritten for a proper fix? Bulia, you mentioned you were going to look at 891852 - did you get a chance to trace down what needs to be done? Rejon, any luck with figuring out the configure issue? Bob, have you been able to get into SourceForge to take a look at the bugs? I can forward you the text of them if not. I think there's 2-3 that hopefully will be quick for you to validate and close for us. Mental, I bet that there's probably a quick fix for 864791. If you won't have time for it today, I might have a chance to look at it tonight or tomorrow night. Finally, would someone mind banging on Inkscape a bunch today looking for crashes, or if there are any features that aren't working? Bryce

2 2

Re: [Inkscape-devel] "dialogs on top" checkbox
by bulia byak 09 Feb '04

09 Feb '04

>It might be worthwhile to disable the menuitem for Windows builds. Indeed! Please do :) BTW I've just checked the latest win build and it seems to work OK, including my yesterday's save fix. And the tutorial and keys reference look acceptable (although they are affected by bug 857316 on my Russian Windows 98). _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

Re: [Inkscape-devel] "dialogs on top" checkbox
by bulia byak 09 Feb '04

09 Feb '04

> > THANKS! So I'm setting the default back to 1. Please add an item to the > > Known Problems in the release notes. > >Will do... I already did this :) Also added about autoraise (still) not working on windows. Guess I'll need to do some windows fixing for the next release. _________________________________________________________________ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

Re: [Inkscape-devel] "dialogs on top" checkbox
by bulia byak 09 Feb '04

09 Feb '04

>Comitted. Sorry for the delay; I need to eat and sleep sometimes too, >you know ^_- > >Fwiw, don't use this code as an example for anything. It's a very ugly >hack that I did just so we'd have something for the release. THANKS! So I'm setting the default back to 1. Please add an item to the Known Problems in the release notes. _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

RE: [Inkscape-devel] Example SVGs + Samples
by bulia byak 09 Feb '04

09 Feb '04

I added one Windows screen from Vellum and two more showing the keys reference on Linux. _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 2

RE: [Inkscape-devel] 0.37rc1
by bulia byak 09 Feb '04

09 Feb '04

OK, the problem is fixed. The new nodepath.cpp is committed. _________________________________________________________________ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

RE: [Inkscape-devel] 0.37rc1
by bulia byak 09 Feb '04

09 Feb '04

Oops. Please hold on releasing. I found a problem in nodepath editing, investigating. >Because I fell asleep I had the unique opportunity to be awake when no >one else was, so there was actually a point where doing repeated 'cvs >update's didn't reveal new code ;) I went ahead and made "release >candidate" which is available here. From reading the backlog, I'm >guessing this isn't final - but I think it's not a bad idea to have. > >http://sourceforge.net/project/showfiles.php?group_id=93438&package_id=9911… > > --Ted > ><< signature.asc >> _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

RE: [Inkscape-devel] 0.37rc1
by bulia byak 09 Feb '04

09 Feb '04

>Because I fell asleep I had the unique opportunity to be awake when no >one else was, so there was actually a point where doing repeated 'cvs >update's didn't reveal new code ;) I went ahead and made "release >candidate" which is available here. From reading the backlog, I'm >guessing this isn't final - but I think it's not a bad idea to have. Great. I don't know if any code changes will be needed, but please don't forget to update the NEWS file from the release notes on Wiki - I updated them today. _________________________________________________________________ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

1 0

Tabs to Spaces
by Bryce Harrington 09 Feb '04

09 Feb '04

One of the tasks for this release was to decide on a tab vs. spaces strategy. This has been a controversial issue and has been a challenge to achieve concensus on, but it sounds like the group mostly favors changing tabs to spaces. A few people still favor tabs, so unfortunately we don't have a 100% concensus as we'd like, but I think we have a rough enough concensus favoring this that we can consider it decided. Before we do, does anyone have a major problem with replacing all the tabs in the codebase with spaces, that we should resolve beforehand? The main issue raised so far has been providing config files or modelines that cause vi and emacs to use spaces instead of tabs, so that there's no extra work for people to adhere to this as a standard. The next two questions are: Does anyone know a good method to untabify the entire codebase in one shot? Does anyone know how best to put modelines in the codebase to get vi and emacs to behave correctly? Bryce

7 7

0.37rc1
by Ted Gould 09 Feb '04

09 Feb '04

Well, Because I fell asleep I had the unique opportunity to be awake when no one else was, so there was actually a point where doing repeated 'cvs update's didn't reveal new code ;) I went ahead and made "release candidate" which is available here. From reading the backlog, I'm guessing this isn't final - but I think it's not a bad idea to have. http://sourceforge.net/project/showfiles.php?group_id=93438&package_id=9911… --Ted

1 0

Re: [Inkscape-devel] back to work
by bulia byak 09 Feb '04

09 Feb '04

>Has anyone had a chance to bang on the app for a while to make sure >there aren't other issues we've overlooked? Actually I've been using it almost daily for the last month. I submitted all problems I find to the tracker (unless I can quickly fix them myself, in which case I do that :) My personal gut feeling is that it's quite stable by now, at least on Linux. I doubt we can discoved any real show-stopper in the last day. _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=dept/features&pgmarket=en-ca&RU=http%3a%2f%2fjoin…

2 1

"dialogs on top" checkbox
by bulia byak 09 Feb '04

09 Feb '04

Mental, as discussed yesterday, please add a checkbox for options.transientpolicy={0|1} to the Dialogs menu today, so that we could set it to 1 in preferences-skeleton.h. I will be very disappointed if we ship the release with 0 as the default. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 2

back to work
by Jonathan Phillips 09 Feb '04

09 Feb '04

Hey all, I just got back to my normal location at UC-San Diego. It is nice to be back in my 24 work zone. I'm thinking of a few things and decided that I would itemize rather than ramble-ize. 1.) PAYPAL: Do we have some form of paypal donations setup for our project. I would like to look into this. I've even been toying with the idea of getting some form of funding from UCSD to plugin to our project to "encourage" certain features to be implemented. I don't know if the "bounties" concept fits, but is interesting. 2.) NODE USABILITY: I don't know what to call it, but I will call it the extra-functionality "handle" for spirals, stars, circles and squares (rounding) is hard to see against a white background. I think we need to do something about the visibility of this. Maybe either by putting a white border around it, or making it reverse on colors. The normal nodes are hard to see sometimes as well. I think we should change the color of the nodes to be darker, or have a white outline or something to pop-them out against black and other dark colors. Just a thought...if you are all interested in this, then I will add another RFE. 3.) TIMETABLE for the next release? I know we have many many many more tasks to complete for this milestone, but now that we are after the new year, maybe we could have a target date for getting the current milestone complete. How about Sunday, February 1st? Cool... Jon -- Jonathan Phillips <jon@...15...>

4 6

RE: [Inkscape-devel] Status Update for Inkscape release work
by bulia byak 09 Feb '04

09 Feb '04

> > Now this function uses document->uri which I build in >sp_document_create() > > from the pathname on the command line and the current dir, using > > inkscape_rel2abs (which even does some normalization of things like .. >in > > the middle). It may probably still fail for some exotic cases, but the >basic > > operations that I tested now work OK. I would like to have this tested on Windows. Bob, could you please do a quick check that the latest CVS saves files properly on Windows, before an after a save-as into a different directory? _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

2 1

Re: [Inkscape-devel] tutorial, terms
by bulia byak 09 Feb '04

09 Feb '04

By the way we need one more term for those blue arrows around a selection. Of course we can use "node handles," "shape handles," and "selection handles" but I think this is wordy and may be confusing. >I'm not happy with control and handle. The terminology I feel is most >common is that the tangent specifiers on bezier curves are called 'control >handles'. I would call everything but the 'nodes' 'control handles'. I >don't see any advantage from separating bezier control handles from >rectangle control handles. > >comments? > >njh _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

Re: [Inkscape-devel] tutorial, terms
by bulia byak 09 Feb '04

09 Feb '04

>I'm not happy with control and handle. The terminology I feel is most >common is that the tangent specifiers on bezier curves are called 'control >handles'. I would call everything but the 'nodes' 'control handles'. I >don't see any advantage from separating bezier control handles from >rectangle control handles. They are different objects with different behavior. If we don't offer standard terminology, people will be inventing their own to differentiate them. If you don't like controls and handles, please propose another variant. _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=dept/features&pgmarket=en-ca&RU=http%3a%2f%2fjoin…

1 0

tutorial, terms
by bulia byak 08 Feb '04

08 Feb '04

The tutorial is switched to Bitstream Vera and slightly updated. The icon which was misaligned should be OK now. Please test. Also I started a wiki page on Inkscape terminology: http://inkscape.org/cgi-bin/wiki.pl?InkscapeTerminology So far it has a few terms that are most often confused. Please argue if you disagree with anything, otherwise please remember these terms and apply them correctly. (That includes myself - I used some of them confusingly before, now it's time to straighten that out.) _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

4 3

RE: [Inkscape-devel] Too big objects crash-win32
by bulia byak 08 Feb '04

08 Feb '04

>Ah. Dimension size. I do not get this problem. >Take an object. Press ctrl+> repeatedly. >Sooner or later it will go to dimensions of W=1 000 000, H=1 000 000 > At this point press ctrl+< takes it back down smoothly to original >size. >Continue ctrl+> repeatedly. >Sooner or later it will go to X=-1 000 000, H=-1 000 000 > At this point press ctrl+< takes it back down smoothly. Perhaps you were not persistive enough :) As I keep ctrl+>ing, on my system the sequence of events is this: - 100000 size is reached and is not updated anymore (this is the limit of the size spinbuttons on the panel); - the object becomes invisible (but its center mark is still visible, and you may be able to see it again if you zoom out); - after about 30 more ctrl+> after the disappearance (no matter how fast you press) Inkscape crashes. Of course it's not what normal users will do in their daily work, but still it must not crash. _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

2 1

RE: [Inkscape-devel] Status Update for Inkscape release work
by bulia byak 08 Feb '04

08 Feb '04

>891852 saves to old location after save-as [ted] >Bulia, you mentioned you were going to look at 891852 - did you get a >chance to trace down what needs to be done? Fixed. The source of the problem was that sp_file_save_document() always used the value of sodipodi:docname - which is supposed to be just name, without directory. Therefore it ALWAYS saved into the current directory. This was wrong not only after save-as but also when you load a file from a directory other than current, e.g. inkscape ../file.svg Now this function uses document->uri which I build in sp_document_create() from the pathname on the command line and the current dir, using inkscape_rel2abs (which even does some normalization of things like .. in the middle). It may probably still fail for some exotic cases, but the basic operations that I tested now work OK. _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

2 1

Example SVGs + Samples
by Jonathan Phillips 08 Feb '04

08 Feb '04

Hello All, In order to get our release out the gate, I would like to make some new screenshots and get some of you to post some sample SVGs into the samples folder in CVS. Please just take like 5 minutes to do this. I'm including a few sample SVGs in the documentation section on the website as well. Screenshots I run Inkscape on Linux, and would appreciate to get some 1024by768 screenshots from non-Linux builds of Inkscape. Please send me a couple of screenshots OS X and win32 folks, or even better, you can add them to the site yourself (http://inkscape.org/cgi-bin/wiki.pl?WebsiteEditing), but I know how time goes, so you can send them to me and I'll format them for the site. Thanks, Jon -- Jonathan Phillips <jon@...15...>

1 0

Re: [Inkscape-devel] Feature freeze 2/7/04 20:00 GMT
by bulia byak 08 Feb '04

08 Feb '04

> > 881397 too big objects crash >How big is too big? Here big means big in dimensions, not in file size. Take any object and press ctrl+> repeatedly - sooner or later it will crash. _________________________________________________________________ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 2

Re: [Inkscape-devel] keys and mouse reference
by bulia byak 08 Feb '04

08 Feb '04

>The tutorial is still using mainly Times New Roman and looks good with this >font. Will fix today. >If Bitstream Vera Sans is required will it be packaged with the Win32 >build? Good idea. Bob? _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=dept/features&pgmarket=en-ca&RU=http%3a%2f%2fjoin…

2 1

keys and mouse reference
by bulia byak 08 Feb '04

08 Feb '04

There's now a Keys and Mouse command in Help. It opens an SVG keyboard reference. (Do "make install" so it is copied to where Inkscape expects to find it). The reference is zoom-based, i.e. you are supposed to first find the section you need using the large section headings; then you zoom into the section and read smaller subsection headings; then you zoom into the subsection and read the keyboard shortcuts; if interested, you can zoom even more to read small-type comments and explanations. I think it's a natural approach for Inkscape with its 256x zoom, and it's pretty convenient. Of course it is assumed that you already know at least how to zoom and pan, but I think it's a safe assumption. The reference uses Bitstream Vera Sans exclusively as suggested by njh, so this is now a required font for Inkscape. I will now change the tutorial to Vera as well. I'm interested to know how they come up on other computers and platforms, if there are any misalignments, etc. Please send a screenshot if you have a problem. The svg file is generated from doc/keys.xml by doc/keys-svg.xsl (I use Saxon 7 but it should work with any processor). I will now write another simple stylesheet to generate the same reference in HTML. Comments welcome as always. _________________________________________________________________ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

3 4

Feature freeze 2/7/04 20:00 GMT
by Bryce Harrington 08 Feb '04

08 Feb '04

I think the time's right to go ahead and feature freeze and put full focus on bug fixing. Let's set the feature freeze time to Saturday 20:00 GMT (that's high noon for those of us in the PST timezone). Here are the remaining code cleanup tasks: ** Review #if 0 code in sp-gradient.cpp, sp-root.cpp [tvon] ** Review #if 0 code in *-chemistry.cpp and seltrans.cpp [pjrm] ** Review #if 0 code in svg/ ** Review #if 0 code in display/ [mental] ** Review #if 0 code in helper/ ** Review #if 0 code in xml/ I haven't seen tvon in a while so perhaps someone else can take his review task? I will post the info on making release packages tomorrow morning. We are down to 9 critical bugs that will need to be addressed prior to our final release: 891852 saves to old location after save-as 850971 Crash while deleting namedview in XML editor 852976 win32: mouse becoming unresponsive 871582 path display freeze [mental] 864791 XML editor: crash when canceling quit [mental] 863373 ./configure not present in nightly tarball snapshot 863371 --gnome-print option 881397 too big objects crash 887428 R&B channels reversed on win32 when printing [ishmal] The ./configure issue looks like just a tweak needed to the nightly build script - Ted or Jon, perhaps you could solve this easily? Ishmal, could you verify that the mouse unresponsive bug on Win32? It's possible this bug is completely fixed now. Also see if you can validate the fix that mental did for 881397, that it works on Win32 too; if so we can mark that one done too. The --gnome-print option bug looks like it might be spurious and not need fixing - could someone try doing `./configure --with-gnome-print` and see if they can reproduce the bug? Jon? There are a couple XML Editor bugs; I looked into each and they did not look as trivial as some of the other bugs. Could someone figure out what we could do to fix these? Mental, do you think 871582 is fixable this weekend? If not, how critical is the bug - could it be postponed for next release? Bryce

4 6

Packagers' info
by Bryce Harrington 08 Feb '04

08 Feb '04

I've updated the Wiki page on CreatingDists with some directions for creating and naming the various kinds of packages we'll need. http://www.inkscape.org/cgi-bin/wiki.pl?CreatingDists Last time we kicked butt at collecting packages for a wide range of platforms. We wish to do the same this time, and improve on this by adopting a standardized naming convention for the packages. Specifics are on the wiki page. Here's examples: inkscape-0.37-1.win32.exe inkscape-0.37-1.i386.deb inkscape-0.37-1.athlon.deb inkscape-0.37-1.ppc.deb inkscape-0.37-1.sparc.deb inkscape-0.37-1.mdk80.i386.rpm inkscape-0.37-1.mdk80.i686.rpm inkscape-0.37-1.mdk91.i386.rpm inkscape-0.37-1.mdk91.i686.rpm inkscape-0.37-1.fc1.i386.rpm inkscape-0.37-1.fc1.i686.rpm inkscape-0.37-1.fc1.athlon.rpm inkscape-0.37-1.fc1.ppc.rpm Employing a standard naming convention will make it clearer for people to find the package they need. In order to put some bounds on the quantity of packages we could accumulate, let's limit the varients of a given distro to two for a given architecture - an "oldish" one for compatibility, and a "newish" one for those on the bleeding edge. Let's also limit ourselves for x86 architectures to just i386 and i686. As a minimum, for announcing the release lets strive to have the following covered: * A win32 package * Debian packages for i386, i686, and ppc * rh73 and rh90 packages for i386, i686, and ppc * mdk92 package for i386 * fc1 package for i386 * suse90 package for i386 * fink package for ppc Based on our prior release I think we should have little problem meeting these base requirements. If we can get more distros and architectures, that's great! Bryce

1 0

keys reference in HTML
by bulia byak 08 Feb '04

08 Feb '04

Is now here: http://inkscape.org/doc/keys.html as well as in doc/ in the distribution. So if you want to change anything, you need to edit doc/keys.xml and regenerate SVG and HTML as described here: http://inkscape.org/cgi-bin/wiki.pl?KeyboardShortcuts _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

Please change the subject line if appropriate.
by Alan Horkan 08 Feb '04

08 Feb '04

Rather than ask anyone person directly I would like to politely request that people try to change the subject line a little more often. I'm finding it extremely confusing to try and follow all these different threads with the same subject lines. (It would probably be easier if did something to break up the list digest and got my mailer to thread everthing nicely but I would still very much appreciate it). Thanks in advance. Replies offlist please if necessary. - Alan PS I first wrote this message Feb 8 but didn't send it as I was hoping that if I concentrated really hard it my just happen due to my physic powers but it seems they have failed me.

2 1

Inkscape Feature Freeze
by Bryce Harrington 07 Feb '04

07 Feb '04

At this time we are temporarily closing the codebase to new features or other significant codebase changes until we are ready to cut the release branch. Until that time, please focus on bug fixing, testing, and packaging/documentation enhancements. Code cleanup work, including removal of the 'if 0' code, conversion of tabs to spaces, cleaning up compiler warnings, etc. should either be postponed or completed soon and with great care and testing. In general I would encourage everyone to set aside feature development or code cleanup work to focus on testing the application and fixing critical bugs. The sooner we can work through this process, the sooner we can have the release completed and be back to regular development. :-) If instead you'd like to continue with development or code cleanup, please do your work on a branch and plan to merge it subsequent to the release. Thanks, Bryce

1 0

Re: [Inkscape-devel] prerelease
by bulia byak 07 Feb '04

07 Feb '04

>Works for me, although I'd really like to see us focus on getting the >official release out as swiftly as possible. Could you put the >prerelease out today? I'm thinking if we put in a good weekend we can >probably get all the outstanding items completed and be ready for an >official release by Sunday evening. We're awfully close. If I am to select only two bugs to be fixed, they are these: http://sourceforge.net/tracker/index.php?func=detail&aid=866197&group_id=93… http://sourceforge.net/tracker/index.php?func=detail&aid=881397&group_id=93… Can anyone please have a look? The rest look less like release blockers, or are too hard to fix at this time. Also there's a nasty bug 891852, but that one I'll try to fix myself today. _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

1 0

prerelease
by MenTaLguY 07 Feb '04

07 Feb '04

Since some folks seem worried about us being "scooped" by the upcoming Sodipodi release... How would you all feel about doing an Inkscape prerelease -- that is, take a known-good CVS snapshot this weekend, make a dist tarball from it, call it .. 0.37-rc1, and put it on the site with something like the usual announcement and release notes. As it's a prerelease, we won't need to worry about filing (all) the rough edges off or packaging for other distros, but we'll still have put our stake in the sand. Maybe that way we could focus better on getting the final release out quickly, without feeling so much pressure to rush. -mental

3 2

Re: [Inkscape-devel] tutorial ready
by bulia byak 07 Feb '04

07 Feb '04

>Good work bulia. I learned more from your help tutorial in five minutes >than >I have in the past few months of fumbling and mumbling. It would be good to >have the entire help tutorial in SVG form. Is this what you have in mind? >This could include detailed examples of more complex actions, how to change >preferences etc.etc. Yes, I will be expanding it, and probably break it into two or three parts (basic/medium/advanced). >I have attached two screenshots. The first is general. The second shows a >small icon misplacement that occurs with my nachine. Let me know if you >want >any specific shots. Thanks for the icon shift report, I'll have to change that because font alignment across systems is obviously unreliable. As for shorts, I wanted to post them on the site. Your Windows shots are even better for that, because they will show off both the tutorial and the windows port. So please send a couple screens showing the entire window with all tools and decorations, with perhaps the first "page" of the tutorial and then with some nice graphics down below. I will then put them on the site (or if you have CVS access you can do this yourself, see http://inkscape.org/cgi-bin/wiki.pl?WebsiteEditing) >https://sourceforge.net/tracker/index.php?func=detail&aid=859364&group_id=9… >I do not get this bug on my machine. Win32. OS2000. build0402030318. I think I have since fixed that mysterious bug. _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 2

about screen
by Jonathan Phillips 07 Feb '04

07 Feb '04

I think something I will do is adapt the nice graphical ledger at the bottom/top of the TUTORIAL for our about window tomorrow. Good job BULIA/NJH and others on that! Jon -- Jonathan Phillips <jon@...15...>

3 3

text undo bug
by MenTaLguY 07 Feb '04

07 Feb '04

Bulia, you were right. Not only was the bug in the undo code, it was in the Repr transaction logging stuff. The problem was that, on reversion of the log, content change events were pulling from the wrong union member, and erroneously "reverting" to the "after" state rather than the proper "before" state. Fixed in CVS. -mental

1 0

Re: [Inkscape-devel] Re: undoing text node changes
by bulia byak 07 Feb '04

07 Feb '04

> > SF is up again. Baically the order of undo actions is wrong, and some > > actions may be lost, when text nodes are edited (no matter whether it's >done > > by text tool or by XML editor). > >As I've already commented, the problem is that there is no unified undo >layer for repr actions and other things (like text cursor movement). > >The Repr transaction logging layer is fine as it is -- the thing is that >we need undo records at the SPDocument layer that can hold more than >just Repr transaction logs. I may be misunderstanding you. What you say may be correct, but it does not apply to this bug as far as I can tell. What needs to be logged here is nothing but a repr change, namely a change in the content of the text node inside tspan. This is what is broken. The fact that undo does not know about cursor movements is irrelevant - it must simply restore the previous content of the node, and that's what it cannot do properly. The fact that the behavior is the same no matter whether you edit the text node by the text tool or directly in the XML editor proves that the text tool is not the problem, the problem is inside the undo system. >I do have a rough sketch in my mind for a new undo abstraction -- >basically a generic undo stack and a 'log entry' base class that variuos >subsystems can subclass for undoing/redoing specific things. > >We could even use it for the next/prev zoom feature (i.e. use one of >these generic undo stacks for document editing history, and another for >zoom history). It's very nice, but currently I'm concerned about this particular bug. I think it must be fixed before the release, because it's pretty annoying. Sometimes (I couldn't figure exactly when) editing text even breaks undo completely, i.e. I get "operation did not complete transaction" and cannot undo any more. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

[Fwd: [Sodipodi-list] Sodipodi 0.34 soon]
by Jonathan Phillips 07 Feb '04

07 Feb '04

-----Forwarded Message----- From: Lauris Kaplinski <lauris@...39...> To: sodipodi-list(a)lists.sourceforge.net Cc: gnome-i18n@...45... Subject: [Sodipodi-list] Sodipodi 0.34 soon Date: Sat, 07 Feb 2004 01:11:38 +0200 Hello! Sodipodi 0.34 is scheduled to be released early next week, so now is good opportunity to update translations. Very little is changed, few wordings here and there and some new strings, so the update should be easy. If you prefer tarball to CVS, the beta package with frozen strings can be downloaded from: http://lauris.kaplinski.com/sodipodi/sodipodi-0.34-beta1.tar.gz Best wishes, Lauris Kaplinski ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Sodipodi-list mailing list Sodipodi-list(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sodipodi-list -- Jonathan Phillips <jon@...15...>

1 0

Re: [Inkscape-devel] inkview slideshow - was fullscreen
by Peter Linnell 07 Feb '04

07 Feb '04

Sources for Open Office XML specs: http://www-106.ibm.com/developerworks/xml/library/x-think15/ http://xml.openoffice.org/

1 0

RE: [Inkscape-devel] fr.po file
by bulia byak 06 Feb '04

06 Feb '04

>I have done an update of the fr.po file. Thanks, but your file seems to be quite old. Recently a lot of code was rearranged, and plenty of new interface string were added. The CVS version of fr.po is 85 Kb (even though many strings are not translated there) versus 76 Kb of your file. Can you update the latest CVS version of the file instead? Or at least, can you merge the new translations you made into the CVS file? http://cvs.sourceforge.net/viewcvs.py/inkscape/inkscape/po/ _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

fr.po file
by Frederic Rodrigo 06 Feb '04

06 Feb '04

Hi, I have done an update of the fr.po file. Fred.

2 1

Re: [Inkscape-devel] drawing tools broken
by bulia byak 06 Feb '04

06 Feb '04

>were working (such as CVS)? I find your lack of faith disturbing - why are >you allowed to make keybinding changes, yet others are not allowed to fix >real problems with the code? I never said you are "not allowed". It's just that I never commit 10+ files at once. (To be honest, I just committed 9, but it's a rare exception, and the changes are similar and minor). And I test thoroughly before committing (although errors do slip in, but they are usually minor or not easy to notice). I'm not saying that you don't test, but it's obviously more difficult to test even the basic functions when you commit lots of changes in lots of files, as has been done recently. >You claim it is severely broken, yet the fix was to add one line. Please >don't over react. "Severely" does not mean "hard to fix". It only means "very visibly, rendering some basic functions unusable". I think it's an accurate description of this bug. Sorry if I sounded rude. I have nothing against cleaning up. I especially appreciate your and Peter's work. But I simply don't like the idea of cleaning up files on schedule, and doing it so close to a release. It is almost inevitable that when lots of files are touched, errors are introduced. I saw (maybe incorrectly) a confirmation of my suspicions in this bug, hence the reaction. _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus&pgmarket=en-ca&RU=http%3a%2f%2fjoi…

3 2

Re: [Inkscape-devel] Remaining Tasks for 0.37 Release
by Robert Crosbie 06 Feb '04

06 Feb '04

Isn't the bonobo stuff not being compiled at all right now? To save the effort, perhaps that could be left out until someone figures out that portion of the code. Cheers, Rob. http://members.rogers.com/rcrosbie >From: Bryce Harrington <bryce@...1...> >To: <inkscape-devel(a)lists.sourceforge.net> >Subject: Re: [Inkscape-devel] Remaining Tasks for 0.37 Release >Date: Thu, 5 Feb 2004 09:48:46 -0800 (PST) > >Here's today's updated list: > > * Review all if 0 code: > * (DONE) Review #if 0 code in desktop*.cpp and svg-view.cpp files >[ted] > * Review #if 0 code in bonobo/ > * Review #if 0 code in sp-gradient.cpp, sp-root.cpp [tvon] > * Review #if 0 code in file.h, interface.cpp, geom.cpp, inkscape.cpp, >and > style.cpp [ishmal] > * Review #if 0 code in *-chemistry.cpp and seltrans.cpp [pjrm] > * Review #if 0 code in svg/ > * Review #if 0 code in display/ [mental] > * Review #if 0 code in xml/ > > * Convert all tabs into spaces in codebase > * dialogs folder [rejon] > * Rename all 'SPFooBar' routines to 'FooBar' and put into > namespaces [mental] > * Update coding-style.txt from the Coding_Style page in wiki > >Sounds like folks think we would be better suited to having humans do >the tab conversion rather than using a script? If so then we could >use more volunteers to help on that; pick a subdir or set of files and >let us know. If you know a trick to doing it with a script, do share! > >Bryce > >On Wed, 4 Feb 2004, Bryce Harrington wrote: > > When all of these are done, we will feature-freeze, and the focus will > > be on bug fixing. There are 10 bugs rated critical, that will need > > fixing. 5 of these do not yet have an owner so we need volunteers to > > take care of them. We also need to review the lower priority bugs to > > see if there are additional critical ones that will need fixing. > > > > Once the critical bugs are fixed, we will establish the stable branch >for > > 0.37 and the release team can get to work on packaging and creation of > > binaries. Once we have binaries for Windows, OSX, and a couple common > > linux distros, we can begin sending out announcements. :-) > > > > Bryce > > > >------------------------------------------------------- >The SF.Net email is sponsored by EclipseCon 2004 >Premiere Conference on Open Tools Development and Integration >See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. >http://www.eclipsecon.org/osdn >_______________________________________________ >Inkscape-devel mailing list >Inkscape-devel(a)lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/inkscape-devel _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

code cleanup
by Jonathan Phillips 06 Feb '04

06 Feb '04

Hello all... I don't think that it will be possible to get our entire codebase up to to the coding style at the current pace, until next week. I think we are going to need to make a conscious decision to attack this untabbifying code systematically (I've only gotten through the dialogs folder in a couple of days!!!), or only untabbify the code for this release (faster), or put this off till next release (and try and get a release out soon--say by weekends end) What do you all think? I understand Bulia's concern about breaking things so close to release and I see how long this takes by hand...(a long time). Would it not be better to knock out the bugs and #if 0 and then put the rest till next release (bug hunt). I think we need to get a stable release up with bool. ops asap. Jon -- Jonathan Phillips <jon@...15...>

3 4

RE: [Inkscape-devel] code cleanup
by bulia byak 06 Feb '04

06 Feb '04

>long time). Would it not be better to knock out the bugs and #if 0 and >then put the rest till next release (bug hunt). Sounds fine to me. Especially fixing bugs. >I think we need to get a >stable release up with bool. ops asap. By the way, Lauris is working on his own version of bool ops. If we're serious about competing with Sodipodi, we must be the first to roll out a release with booleans. (By the way, I noticed that Sodipodi development has certainly sped up since the birth of Inkscape - that's how competition works.) _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 2

drawing tools broken
by bulia byak 06 Feb '04

06 Feb '04

I think this frenzy of replacements and cleanups should have been done after a release, not now. It's dangerous. Right now Inkscape is severely broken: try to draw a rect or ellipse or spiral - they always start from coordinate origin. This is a visible bug, but I'm terrified to think how many more hidden ones were introduced by all this activity. And we wanted to do a release soon. Please someone fix the drawing tools problem ASAP! _________________________________________________________________ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

3 2

Re: undoing text node changes
by bulia byak 06 Feb '04

06 Feb '04

>Well, as a matter of principle SPXMLViewContent shouldn't have to know >anything about desktops -- the calls to sp_desktop_done() should be >happening in src/dialogs/xml-tree.cpp. Not desktop, but document! Surely it must know its document - if it's editing this document. So what's the best way to figure out the document? Can we -> our way from a repr to its document? Or can we pass the SPDocument pointer in a data attached to the dialog or widget? > > 2. Now that the text node changes are undoable, undoing editing a text > > object in XML editor has the same problem as in this bug: > > > > >http://sourceforge.net/tracker/index.php?func=detail&aid=861462&group_id=93… > > > > That is, if you create a text node, switch to any other tool, and then > > reproduce this bug's sequence but using XML editor instead of F8, the >result > > is the same. Which proves that the source of the problem is not text >tool, > > but the wrong way text node changes are logged and undone by the repr > > logger. It was only obvious in text tool because no other tools >create/edit > > text nodes. Can you please have a look into it? > >Sourceforge is down again, so I can't get to the bug tracker. Could you >please summarize the bug for me in more detail? SF is up again. Baically the order of undo actions is wrong, and some actions may be lost, when text nodes are edited (no matter whether it's done by text tool or by XML editor). > >-mental ><< signature.asc >> _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

2 1

Re: [Inkscape-devel] Remaining Tasks for 0.37 Release
by bulia byak 06 Feb '04

06 Feb '04

>problem but I didn't file a report yet So please do, so we don't forget about it. _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0

proposal: text spacing keys
by bulia byak 05 Feb '04

05 Feb '04

Now that the letterspacing support is in, the next thing I will do is the proper support for dx and dy attributes (it's sort-of supported now but actually useless, because only single values are supported, not lists as per SVG spec). This will allow kerning pairs of letters in a string as well as raising/lowering of characters on the string. With all this, I need a set of keys for doing all the space adjustments in a text object from keyboard. These keys must be as consistent as possible with the rest of Inkscape. So, when you are editing text on the canvas: Alt+<, Alt+> expand/contract current line (not the entire text object - I think it's more useful this way) by 1 pixel by adjusting letter spacing Shift+Alt+<, Shift+Alt+> same, by 10 pixels (The above keys are already implemented - you can test them) Alt+[, Alt+] expand/contract text object vertically by 1 pixel by adjusting line spacing Shift+Alt+[, Shift+Alt+] same, by 10 pixels Alt+left, Alt+right kern closer/farther the pair of characters before and after the cursor by 1 pixel Shift+Alt+left, Shift+Alt+right same, by 10 pixels Alt+up, Alt+down raise/lower the character after cursor by 1 pixel Shift+Alt+up, Shift+Alt+down same, by 10 pixels All these movements are in pixels (as always with Alt), so the actual distance depends on zoom level. Zoom in for finer movement. Suggestions, comments? _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=dept/bcomm&pgmarket=en-ca&RU=http%3a%2f%2fjoin.ms…

1 0