Activity for DjVuLibre
-
Gal Oman created ticket #109
Windows for ARM release
-
Barak A. Pearlmutter committed [b2ea8c] on Djvulibre-git
rrr now defaults to no
-
Barak A. Pearlmutter committed [948d45] on Djvulibre-git
log merge
-
Barak A. Pearlmutter committed [d6464b] on Djvulibre-git
Merge remote-tracking branch 'upstream/master' into debian
-
Dr Bill C Riemers posted a comment on ticket #364
In this case though, I verified the presense of the respective header files and typedef. So it likely a cmake/automake problem. Which I'm sure you already found.
-
For future reference, I had a very similar problem on my Mac for a bit new code I wrote on ARM I was trying to compile on my Mac. It turns out most of those modern C++ definitions are not in default include file set. My solution was just to install G++, and and the instructions for doing so to my README.md file, rather than to make my code explicitly typedef values like this.
-
Mikhail Morozov created ticket #366
djvm hangs when running multiple times in a row
-
cpaldjvu produce broken images
-
Leon Bottou committed [32645a] on Djview-git
djvu mdimporter: simplified schema and info.plist
-
Leon Bottou committed [80389f] on Djview-git
djvu mdimporter: simplified schema
-
DjVuLibre released /DjView/4.12/djview-4.12.3.tar.gz
-
Leon Bottou modified ticket #364
Invalid typedef in GString.h breaks C++ builds with modern compilers
-
Fixed
-
Added missing semicolon to fix #364
-
Make new script executable
-
Support for small bundles (e.g. built within homebrew)
-
Correct last commit (reversed checks once again.)
-
More Fedora patches (parameter validation avoid zdiv exceptions)
-
-
Janusz posted a comment on ticket #364
Since several days AskTheCode is unable to compile a simple program, prepared by itself, namely has_composite_shapes.cpp. The repository is now quite a mess after chaotic experiments: https://github.com/jsbien/labelshapes. FYI, this is the link to the chat https://chatgpt.com/share/686e7fc0-0f14-800d-985b-dee2719899ab; the last advice also doesn't work.
-
Yes. But this line is only activated when neither HAVE_STDINT_H, HAVE_INTTYPES_H, not _WIN32 are defined. These days, all systems have <stdint.h> and C++ specifies <cstdint> (which would have been so much simpler if these things existed when djvulibre was written.). So I am curious about your compilation setup. - Leon</cstdint></stdint.h> if HAVE_STDINT_H include <stdint.h></stdint.h> elif HAVE_INTTYPES_H include <inttypes.h></inttypes.h> else ifdef _WIN32 On Jul 9, 2025, at 10:01 AM, Janusz jsbien@users.sourceforge.net...
-
Invalid typedef in GString.h breaks C++ builds with modern compilers
-
Barak A. Pearlmutter committed [495b88] on Djvulibre-git
lintian
-
Barak A. Pearlmutter committed [f61d54] on Djvulibre-git
update patches and log changes
-
Barak A. Pearlmutter committed [e281b2] on Djvulibre-git
Merge tag 'release.3.5.29' into debian
-
Salvatore Bonaccorso committed [b55ebd] on Djvulibre-git
Merge tag 'debian/3.5.28-2.1' into debian
-
Salvatore Bonaccorso committed [80395d] on Djvulibre-git
Prepare changelog for release
-
Salvatore Bonaccorso committed [13e180] on Djvulibre-git
Fix potential buffer overflow in MMRDecoder (CVE-2025-53367)
-
Alexandre Detiste committed [1fba21] on Djvulibre-git
Merge branch 'scrub-obsolete' into 'debian'
-
Debian Janitor committed [f6e18b] on Djvulibre-git
Remove constraints unnecessary since buster (oldstable)
-
Barak A. Pearlmutter committed [b5c2aa] on Djvulibre-git
approved way of snarfing DEB_HOST_MULTARCH
-
Barak A. Pearlmutter committed [1b3abd] on Djvulibre-git
Update standards version to 4.6.2, no changes needed.
-
Barak A. Pearlmutter committed [646e08] on Djvulibre-git
Merge branch 'master' into debian
-
Jelmer Vernooij committed [58afe6] on Djvulibre-git
Merge branch 'lintian-fixes' into 'debian'
-
Update standards version to 4.6.1, no changes needed.
-
Use secure URI in Homepage field.
-
Barak A. Pearlmutter committed [8a7c72] on Djvulibre-git
bump policy
-
Barak A. Pearlmutter committed [d0b5e1] on Djvulibre-git
Merge remote-tracking branch 'upstream/master' into debian
-
Leon Bottou committed [680866] on Djview-git
Added installation instructions
-
Leon Bottou committed [002c2c] on Djview-git
good for qt-6.2.9
-
-
-
Leon Bottou committed [4a285e] on Djvulibre-git
MMRDecoder: more margin in lineruns
-
release 3.5.29
-
-
Leon Bottou committed [bc1449] on Djvulibre-git
Release 3.5.29 (bug fix release)
-
Leon Bottou committed [33f645] on Djvulibre-git
Fix potential buffer overflow in MMRDecoder
-
Increased frequency of crashes on macOS Sonoma
-
Wikipedia sample DJVU file cannot be decoded
-
no batch conversation possible?
-
compiling the ghostscript driver with recent version fails ("missing contrib.mak")
-
Make ddjvu help text mention pdf
-
I would have to see the file a.djvu to understand. Maybe you can try to give the output of command djvudump a.djvu to get an idea.
-
anon created ticket #363
djvups does not preserve the text encoding
-
https://github.com/jwilk-archive/djvusmooth
-
Robert H Risch posted a comment on discussion DjVuLibre Help
Is there a way to add comments and annotations to a djvu file with DjVuLibre? If not, is there another djvu reader that allows this? Thanks in advance.
-
Stephen Jones posted a comment on discussion DjVuLibre Help
Recently I have not been able to access Any2DjVu, the free online site that allowed one to encode various formats, such as PDF, PBM and JPEG, as DjVu documents. Has the site been removed? I could not find any information here about this topic. The site provided results similar to the commercial DjVu encoder. I think I used the site as recently as a few months ago. Thank you.
-
Havlicek stole the ball posted a comment on ticket #362
That seems to have helped. Thank you!
-
I pushed into git a new version of djvudigital that works better with filenames containing spaces.
-
Please use the git repository instead of the released files. $ git clone https://git.code.sf.net/p/djvu/gsdjvu-git djvulibre-gsdjvu Does it work better. Note that I recommend gs-9.55 rather than the new gs-10 versions, for now.
-
fixed djvudigital space issues
-
Fixed djvudigital space issues
-
Fix space problems in djvudigital
-
I've accidentally posted this in the djview section - perhaps this should be moved to the gsdjvu section?
-
Hey, perhaps someone can explain how I can convert pdf files with djvudigital limited to greyscales/monochrome? Cheers
-
Hey :-) As the title says: Is it possible to limit the conversion with djvudigital to certain pages of the input file? Like from page 10 to 20? I know this is possible with pdf2djvu but I couldn't find an option in the djvudigital manual. Cheers
-
no batch conversation possible?
-
compiling the ghostscript driver with recent version fails ("missing contrib.mak")
-
The error message was not very clear either.
-
This points to a corrupted file. Not a very good error message! It denotes a parsing error while reading the djvu file chunk structure. When one writes an incorrect chunk structure, "not ready for writing data" makes some sense. The same messahe when reading an incorrect chunk structure was not a good choice.
-
Kevin Routley posted a comment on ticket #360
How embarressing! My apologies for the error and distraction! I relied on the user report without digging further. The sample DJVU file they provided, and claimed came from that link, is completely bogus when examined.
-
Image rendering differs between Qt5 and Qt6 builds.
-
When the Qt6 reports that the screen implements high dpi scaling ("retina"-style displays), djview , djview generates an image with a higher resolution that matches the true number of pixels and uses this by bypassing the high dpi scaling code. This gives a higher quality image. For me, this works as expected. The QT_ENABLE_HIGHDPI_SCALING=0 variable disables all this behavior, so that Qt6 matches Qt5's lower quality image. This could be useful to save cycles or memory.
-
DJview4 won't open djvu document online
-
It's interesting that wget can download it as text/plain without problem You got lucky. According to the standard, content of mimetype text/plain should encode line breaks as CRLF. So a web server running on Unix might very well convert all LFs to CRLF on the fly. On the other hand, a web server running on Windows would not do this. I rather have djview complain in all cases rather than show a mysterious bug later.
-
Not the same experience here. Things seem to work well under both Sonoma and Sequoia. Short of a reproducible crash, this is hard to debug.
-
.pc file does not work on MinGW
-
The mask extracted with ddjvu is not binary, but greyscale
-
djvm -l cannot handle output from cjb2
-
Cannot reproduce. Works here. Attaching the downloaded file for comparison.
-
Wikipedia sample DJVU file cannot be decoded
-
I'm using DjVuLibre 3.5.28 in my Windows Explorer plugin DarkThumbs (https://github.com/fire-eggs/DarkThumbs). It's working pretty well for me, but a user reported that the sample DJVU file from Wikipedia cannot be handled. Here is the link: https://commons.m.wikimedia.org/wiki/File:Example.djvu DjView 4.12 reports "IFFByteStream not ready for reading chunk".
-
I would have to see your original myfile.djvu.
-
b'Leon Bottou committed [b0ff6f] on Djvulibre-git
Merge /u/trufanov/djvu/ branch master into master
-
Fix initial zoom value description
-
Well, a djvu file can be several things: a single page file (FORM:DJVU). Cjb2 and c44 produce such files. a bundled document (FORM:DJVM) which may contain one of more pages, themselves encoded as FORM:DJVU. It also contain a page directory, and may contain shape dictionaries and navigation menus. Such documents are created with "djvm -c". You can use the program 'djvudump' to know what's inside a djvu file. Command "djvm -l" does not work on the output of cjb2 because it is designed to work on biundled...
-
free Buju posted a comment on ticket #359
A workaround seems to be to run djvm -c bundle.djvu sample.djvu sample.djvu, then djvm -l works on that. From there, it appears superficially to not be too useful. But it actually helped me discover a hack to work around bug 358 (https://sourceforge.net/p/djvu/bugs/358/). djvm -l shows us that page names are assigned as the original filename. These can be used as a handle to make djvused work. I’ll elaborate further in bug 358.
-
John Harry posted a comment on discussion Open Discussion
In my personal view, the DjVuLibre team need not be concerned about this issue, as DjVuLibre should not be licensed under the GPL site. When promoting the use of a standard file format, it is unwise to apply the GPL to the code necessary for handling that format.
-
I would be very happy to be able to extract easily graphic snippets (say in png format) identified by a possibly long list of URLs like this Hebrew(?) string in an old dictionary (for the purpose of OCR). The most elegant way would be to extend ddjvu to handle pagespec not only in the form of page numbers, but also in the form of full URLs (for simplification to local files only). I guess requesting such an extensions would be unrealistic, so I would appreciate any suggestions to solve the problem....
-
Vladimir posted a comment on discussion DjVuLibre Help
It seems that this file should have been automatically generated during the compilation, like all the other "ui_*" files. But, because it is used before it is generated, there is such an error. To solve this problem, I had to find and download this "debug-info" package: http://arm.koji.fedoraproject.org/koji/buildinfo?buildID=144684 http://arm.koji.fedoraproject.org//packages/djview4/4.9/2.fc20/armv7hl/djview4-debuginfo-4.9-2.fc20.armv7hl.rpm Then, I had to extract the "rpm" archive with rpm2cpio...
-
I have described the problem in details at http://teksty.klf.uw.edu.pl/11/ and provided the problematic files. When executing a script djvused doesn't issue any error messages, but in the resulting document text layer hierarchy is corrupt. The script is generated by ocrodjvu, so perhaps it is a culprit: cf. https://bitbucket.org/jwilk/ocrodjvu/issue/12/ocrodjvu-creates-an-incorrect-djvused. However, if the script is incorrect, I would like to know what is exactly wrong. Perhaps it is just too big?...
-
I've created rather sophisticated bookmarks for the dictionary at http://korpusy.klf.uw.edu.pl/djvus/linde-t/01/index.djvu. They work as intended on Linux with djview4 as the browser plugin, but on Windows the Caminova (now DjVu Universe?) plugin is unable to handle them correctly. Is this the Caminova plugin bug? Is there a better forum to ask about the Caminova plugin? Anyway, can the problem be somehow circumvented? Most of the reders of the dictionary use Windows and djview4 doesn't work as a...
-
I come back to this old topic because I've just noticed that the Caminowa plugin cannot handle the outline at http://korpusy.klf.uw.edu.pl/djvus/linde-t/01/index.djvu? What is the reason the djview4 doesn't work as a browser plugin? Is coding it so difficult or time-consuming?
-
ddjvu can indeed generate a pdf (essentially an encapsulated tiff).
-
djvm -l cannot handle output from cjb2
-
djvused -f crashes when feeding it it’s own output (per the man page example)
-
The problem appeared not related to the file format, it is now fixed.
-
I was too optimistic, the impression the program accepts ddjvu produced TIFF as binary graphics was an illusion due to some mistake of mine. I try to persuade ChatGPT to correct the script (https://github.com/jsbien/tmp), we'll see what happens.
-
Thank you very much for your patience and the detailed explanation! My problems were caused by the conversion steps. I will follow your suggestion and use the TIFF output. To my pleasant surprise the next program in the pipeline accepts TIFFs as the input, so conversion is not needed. FYI, I intend to process font tables uploaded to https://github.com/jsbien/early_fonts_inventory/tree/main/font_tables/oDjvu.I need binary files for processing and , instead of running some binarization tool, I prefer...
-
Option -1 or --subsample=1 ensure that the output resolution matches that of the input image. Note that this is the default when you use the --format option (which you should always use.) Note that when you use --format=pbm, you force the output to be a pbm file which can only encode binary images. Even if you were to subsample the image (making the mask gray level), it would be tresholded into a binary image to produce a pbm file. These PBM files start with the two letters "P4". If this is the case,...
1 >
