Comments for CE's Blog http://www.christian-etter.de Stuff I found noteworthy. Fri, 21 May 2010 11:15:11 +0000 hourly 1 http://wordpress.org/?v=3.0.1 Comment on Testing for possible Unicode – ANSI code page compatibility by Anonymous http://www.christian-etter.de/?p=51&cpage=1#comment-699 Anonymous Fri, 21 May 2010 11:15:11 +0000 http://winweb319.win-www.df-webhosting.de/?p=51#comment-699 Thank you for the tip, Christian. I convert the buffer i read to UTF-16 and now i can see correct chars :-) Thank you for the tip, Christian.
I convert the buffer i read to UTF-16 and now i can see correct chars :-)

]]> Comment on Testing for possible Unicode – ANSI code page compatibility by Christian Etter http://www.christian-etter.de/?p=51&cpage=1#comment-696 Christian Etter Fri, 21 May 2010 10:13:36 +0000 http://winweb319.win-www.df-webhosting.de/?p=51#comment-696 I don't know if your text reading api supports UTF8 directly, in case it doesn't you should be able to read the file as binary data and then convert it into UTF-16 on the Windows platform. This can be done by using the MultiByteToWideChar() API. - Christian I don’t know if your text reading api supports UTF8 directly, in case it doesn’t you should be able to read the file as binary data and then convert it into UTF-16 on the Windows platform. This can be done by using the MultiByteToWideChar() API.

- Christian

]]> Comment on Testing for possible Unicode – ANSI code page compatibility by Jean http://www.christian-etter.de/?p=51&cpage=1#comment-693 Jean Fri, 21 May 2010 05:45:17 +0000 http://winweb319.win-www.df-webhosting.de/?p=51#comment-693 Yes, the text file is correct :-) But i need to read it, i tried fread, fgets, fgetws. And i still get uncorrect characters Yes, the text file is correct :-)
But i need to read it, i tried fread, fgets, fgetws.
And i still get uncorrect characters

]]> Comment on Testing for possible Unicode – ANSI code page compatibility by Christian Etter http://www.christian-etter.de/?p=51&cpage=1#comment-684 Christian Etter Thu, 20 May 2010 19:47:46 +0000 http://winweb319.win-www.df-webhosting.de/?p=51#comment-684 Jean, Text is in UTF-8 format, it reads out fine. Try: exiftool -location IMG_0650_web.jpg>location.txt then open the text file with a UTF-8 aware editor. "St. Isaac's Cathedral (Исаакиевский Собор)" This should solve the problem. Christian Jean,
Text is in UTF-8 format, it reads out fine. Try:

exiftool -location IMG_0650_web.jpg>location.txt

then open the text file with a UTF-8 aware editor. “St. Isaac’s Cathedral (Исаакиевский Собор)” This should solve the problem.

Christian

]]> Comment on Testing for possible Unicode – ANSI code page compatibility by Jean http://www.christian-etter.de/?p=51&cpage=1#comment-680 Jean Thu, 20 May 2010 17:08:41 +0000 http://winweb319.win-www.df-webhosting.de/?p=51#comment-680 yes, the output is made to a text file, i read it with Notepad (Notepad reads perfectly those characters). the metadata should be corrected written, i can read them with Imagemagick (i have joined the file in the Exiftool forum) yes, the output is made to a text file, i read it with Notepad (Notepad reads perfectly those characters).
the metadata should be corrected written, i can read them with Imagemagick

(i have joined the file in the Exiftool forum)

]]> Comment on Testing for possible Unicode – ANSI code page compatibility by Christian Etter http://www.christian-etter.de/?p=51&cpage=1#comment-679 Christian Etter Thu, 20 May 2010 16:21:45 +0000 http://winweb319.win-www.df-webhosting.de/?p=51#comment-679 Again, did you output this data to a text file and open it as UTF-8 in a suitable text editor? Is there a chance the data has been written incorrectly? Many third party programs do not handle UTF-8 IPTC correctly. They might just use the system default code page. Please verify by writing cyrillic text using the -@ argfile of ExifTool. - Christian Again, did you output this data to a text file and open it as UTF-8 in a suitable text editor?
Is there a chance the data has been written incorrectly?
Many third party programs do not handle UTF-8 IPTC correctly. They might just use the system default code page. Please verify by writing cyrillic text using the -@ argfile of ExifTool.

- Christian

]]> Comment on Testing for possible Unicode – ANSI code page compatibility by Jean http://www.christian-etter.de/?p=51&cpage=1#comment-678 Jean Thu, 20 May 2010 15:15:57 +0000 http://winweb319.win-www.df-webhosting.de/?p=51#comment-678 Christian Thank you for your answer, i did not know that 'type' command, it works fine. Now i rename/copy the files to temp and Exiftool can read them. But some infos are written in cyrillic (see for example Keywords and By-line) and are not read by Exiftool; it seems to be the normal behaviour due to Perl ? Current IPTC Digest : 9a42ed6cec8fd5344ec946c1cb15501e Application Record Version : 2 Keywords : Àíäðåé, Èâàí, Òàðàñ By-line : Ïåòðîâ Ïåòð IPTC Digest : 9a42ed6cec8fd5344ec946c1cb15501e Image Width : 1704 Image Height : 2272 Encoding Process : Baseline DCT, Huffman coding Christian
Thank you for your answer, i did not know that ‘type’ command, it works fine.
Now i rename/copy the files to temp and Exiftool can read them.
But some infos are written in cyrillic (see for example Keywords and By-line) and are not read by Exiftool; it seems to be the normal behaviour due to Perl ?

Current IPTC Digest : 9a42ed6cec8fd5344ec946c1cb15501e
Application Record Version : 2
Keywords : Àíäðåé, Èâàí, Òàðàñ
By-line : Ïåòðîâ Ïåòð
IPTC Digest : 9a42ed6cec8fd5344ec946c1cb15501e
Image Width : 1704
Image Height : 2272
Encoding Process : Baseline DCT, Huffman coding

]]> Comment on Testing for possible Unicode – ANSI code page compatibility by Christian Etter http://www.christian-etter.de/?p=51&cpage=1#comment-665 Christian Etter Thu, 20 May 2010 08:12:22 +0000 http://winweb319.win-www.df-webhosting.de/?p=51#comment-665 Jean, the bahaviour you are describing is due to a limitation of the Perl interpreter, which can only handle file names which can be represented in the current ANSI code page of your Windows system. Most likely you are using code page 1252, which does not contain cyrillic letters. Basically you have the following options: <ol> <li>Change your system code page to one that supports cyrillic. This will have many side effects and might impact your ability to open files with western accented characters.</li> <li>Use the windows command line to rename the file including the folder name to one that can be safely represented in your code page. After you are done with ExifTool, rename it and move it to the original location. This is a complicated but safe approach if you are planning to modify the file.</li> <li>Do not provide a file name to ExifTool and tell it to read data from stdin instead. Then use the <em>type </em>command to read the file. E.g. <strong>type Кирилица.jpg | exiftool -</strong>. This is guaranteed to work in all cases, but it is difficult to use the same technique for writing.</li> </ol> Jean,
the bahaviour you are describing is due to a limitation of the Perl interpreter, which can only handle file names which can be represented in the current ANSI code page of your Windows system. Most likely you are using code page 1252, which does not contain cyrillic letters. Basically you have the following options:

  1. Change your system code page to one that supports cyrillic. This will have many side effects and might impact your ability to open files with western accented characters.
  2. Use the windows command line to rename the file including the folder name to one that can be safely represented in your code page. After you are done with ExifTool, rename it and move it to the original location. This is a complicated but safe approach if you are planning to modify the file.
  3. Do not provide a file name to ExifTool and tell it to read data from stdin instead. Then use the type command to read the file. E.g. type Кирилица.jpg | exiftool -. This is guaranteed to work in all cases, but it is difficult to use the same technique for writing.
]]> Comment on Testing for possible Unicode – ANSI code page compatibility by Jean http://www.christian-etter.de/?p=51&cpage=1#comment-658 Jean Thu, 20 May 2010 06:33:09 +0000 http://winweb319.win-www.df-webhosting.de/?p=51#comment-658 Hello I use ExifTool under Window (XP, 7). I create a .bat to be able to get the metadata in a .txt with the following syntax: (the .bat is created with WCHAR characters) exiftool.exe -EXIF:All myfile.jpg >myfile.txt It works perfectly with 'western' file names but as soon as i use, for example, cyrillic file names, nothing's happening. Hello
I use ExifTool under Window (XP, 7).
I create a .bat to be able to get the metadata in a .txt with the following syntax:
(the .bat is created with WCHAR characters)

exiftool.exe -EXIF:All myfile.jpg >myfile.txt

It works perfectly with ‘western’ file names but as soon as i use, for example, cyrillic file names, nothing’s happening.

]]> Comment on ExifTool Performance Benchmark by Christian Etter http://www.christian-etter.de/?p=458&cpage=1#comment-529 Christian Etter Thu, 22 Apr 2010 07:25:19 +0000 http://www.christian-etter.de/?p=458#comment-529 Post has been updated. Thanks Thomas. Post has been updated. Thanks Thomas.

]]>