Windows build should be fixed in r543. I did not recognize any problem regarding leptonica (but I just run few OCR on my test images) ;-)
Zd. On Tue, Nov 30, 2010 at 2:37 AM, Ray Smith <[email protected]> wrote: > My merge with the latest Google code is now complete and committed. > > The svn autotools are currently horribly broken for me. (Using make dist > and then trying to build from the tar.gz distribution) I had to make the > following patches in order for it to build, but when it did, it worked: > > * make whines about missing .Plo files and missing .Po files in .libs/* I > had to copy them from my earlier version of 3.01, in which they were all > created by make. I suspect this is a problem with the gettext system. I > built my makefiles with no options to runautoconf and configure on linux > Lucid. > > * libtool does not exist in the default distribution. I copied that from my > earlier version of 3.01. > > This version will not compile with any known version of leptonica! Only > 1.67 and above are compatible at the source level, but the distribution of > 1.67 builds a .so.0 which tesseract fails to find, even after removing the > apt-get version of 1.64. leptonica 1.68 will be out soon to fix this > problem, but in the mean time, I am uploading a .deb package of liblept 1.67 > that outputs .so.1. To fix this temporarily there is a couple of debian > packages in a debian directory that can be used to build on 64 bit linux > systems. I may fix this better tomorrow by removing the dependency on the > function that needs 1.67. This probably also breaks the Windows build. > > Ray. > > > > On Sun, Nov 28, 2010 at 4:21 AM, zdenko podobny <[email protected]> wrote: > >> Just notice - if somebody did not recognize it yet: >> >> in svn (http://code.google.com/p/tesseract-ocr/source/checkout revision >> 527) there is 3.01 code that was build successfully on linux (Mandrivalinux >> cooker 64bit) and Windows (XP SP3, VC++2008 Express). There is info about >> additional 3.01 code coming from Ray in (near) future. >> >> So please try it on other platforms/systems and report problem/submit >> patches in http://code.google.com/p/tesseract-ocr/issues/list in Issue). >> >> If you are willing to create C wrapper (see >> http://code.google.com/p/tesseract-ocr/issues/detail?id=386<http://code.google.com/p/tesseract-ocr/issues/detail?id=386&sort=-id> >> , >> http://code.google.com/p/tesseract-ocr/issues/detail?id=362<http://code.google.com/p/tesseract-ocr/issues/detail?id=362&sort=-id> >> , >> http://groups.google.com/group/tesseract-dev/browse_thread/thread/a348e5a6dbade5d7) >> this could be good time for first version ;-) so it can become part of 3.01 >> final code. >> >> Zd. >> >> On Fri, Oct 1, 2010 at 11:26 PM, Jimmy O'Regan <[email protected]> wrote: >> >>> I've put the code of 3.01 on GitHub - >>> http://github.com/jimregan/tesseract-ocr. I'd intended to push the >>> merge into SVN today, but stupidly used the http address when building >>> the git repository instead of the https address, so I can't push back >>> directly without rewriting the references. That might be for the best >>> though, as I think it might be worth leaving 3.00 as is for a week or >>> two, before pushing out the update. >>> >>> There might still be a few glitches in the build system. >>> >>> -- >>> <Leftmost> jimregan, that's because deep inside you, you are evil. >>> <Leftmost> Also not-so-deep inside you. >>> >> >> > -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/tesseract-ocr?hl=en.
