Sponsored Content
Top Forums UNIX for Advanced & Expert Users Regular expression / regex substition on Unicode text Post 302391747 by thomas.hedden on Tuesday 2nd of February 2010 10:51:44 PM
Old 02-02-2010
Regular expression / regex substition on Unicode text

I have a large file encoded in Unicode that I need to convert to CSV. In general, I know how to do this by regular expression substitutions using sed or Perl, but one problem I am having is that I need to put a quotation mark at the end of each line to protect the last field. The usual regex substitution ...
s/$/"/
... works fine for 7-bit ASCII text, but when I run this on my Unicode text file, the double quotation mark appears at the BEGINNING of the FOLLOWING line, not at the end of the line on which it's supposed to appear.
The file came from a Windows system, but piping through dos2unix doesn't seem to make any difference. I've tried the "use Encode;" pragma with several different encodings, but I get the same result. Perhaps I'm doing something wrong. Does anyone know of a special library function intended for this purpose, a Perl pragma, etc., that would accomplish this easily? This should be a trivial problem.
Thanks in advance for any suggestions.
Tom

---------- Post updated at 10:51 PM ---------- Previous update was at 10:49 AM ----------

As I mentioned, the regex ...
s/$/"/
... puts the `"' at the beginning of the following line, and piping through dos2unix doesn't matter one way or the other. Using `\n' instead of `$' doesn't
make any difference. However, I discovered an interesting fact: The regex ... s/\r/"/ # use `\r' instead of `$' or `\n' ... gives the expected result, and does so without piping through dos2unix making any difference! I also found that a C program using a wchar_t declaration behaves similarly. That is, a character that appears as if it should be output BEFORE the EOL actually appears after it, if it is matched as '\n', but if it is matched as '\r' then it is output as expected.
My immediate problem is solved, however I wonder whether this doesn't show a bug in Perl or in its regular expression engine ...
This seems to be a clear case where Unicode text is handled differently than non-Unicode text.
Any opinions?
Tom

Last edited by pludi; 02-03-2010 at 02:29 AM.. Reason: text format, rm URL, and another text format
 

10 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

Regular Expression + Aritmetical Expression

Is it possible to combine a regular expression with a aritmetical expression? For example, taking a 8-numbers caracter sequece and casting each output of a grep, comparing to a constant. THX! (2 Replies)
Discussion started by: Z0mby
2 Replies

2. Linux

Regular expression to extract "y" from "abc/x.y.z" .... i need regular expression

Regular expression to extract "y" from "abc/x.y.z" (2 Replies)
Discussion started by: rag84dec
2 Replies

3. Shell Programming and Scripting

Regular expression (regex) required

I want to block all special characters except alphanumerics.. and "."(dot ) character currently am using // I want to even block only single dot or multiple dots.. ex: . or .............. should be blocked. please provide me the reg ex. ---------- Post updated at 05:11 AM... (10 Replies)
Discussion started by: shams11
10 Replies

4. Shell Programming and Scripting

Integer expression expected: with regular expression

CA_RELEASE has a value of 6. I need to check if that this is a numeric value. if not error. source $CA_VERSION_DATA if * ] then echo "CA_RELESE $CA_RELEASE is invalid" exit -1 fi + source /etc/ncgl/ca_version_data ++ CA_PRODUCT_ID=samxts ++ CA_RELEASE=6 ++ CA_WEEK_NO=7 ++... (3 Replies)
Discussion started by: ketkee1985
3 Replies

5. Shell Programming and Scripting

How can I get the matched text when using regular expression.

Hello: (exp) : match "exp",the matched text is stored in auto named arrays. How can I get the matched text ? What is the name of the auto named arrays on linux shell ? (4 Replies)
Discussion started by: 915086731
4 Replies

6. Programming

Perl: How to read from a file, do regular expression and then replace the found regular expression

Hi all, How am I read a file, find the match regular expression and overwrite to the same files. open DESTINATION_FILE, "<tmptravl.dat" or die "tmptravl.dat"; open NEW_DESTINATION_FILE, ">new_tmptravl.dat" or die "new_tmptravl.dat"; while (<DESTINATION_FILE>) { # print... (1 Reply)
Discussion started by: jessy83
1 Replies

7. Shell Programming and Scripting

passing a regex as variable to awk and using that as regular expression for search

Hi All, I have a sftp session log where I am transferring multi files by issuing "mput abc*.dat". The contents of the logfile is below - ################################################# Connecting to 10.75.112.194... Changing to: /home/dasd9x/testing1 sftp> mput abc*.dat Uploading... (7 Replies)
Discussion started by: k_bijitesh
7 Replies

8. Emergency UNIX and Linux Support

Regular expression (regex) clean up text

Hi, Server - MEDIAWIKI - MYSQL - CENTOS 5 - PHP5 I have a database import of close to a million pages into my wiki, mediawiki site, the format that were left with is not pretty, and I need to find a way to clean this up and present it nicely... I think regex is the best option as I can... (1 Reply)
Discussion started by: lawstudent
1 Replies

9. Shell Programming and Scripting

regular expression with shell script to extract data out of a text file

hi i am trying to extract some specific data out of a text file using regular expressions with shell script that is using a multiline grep .. and the tool i am using is pcregrep so that i can get compatibility with perl's regular expressions for a sample data like this, i am trying to grab... (6 Replies)
Discussion started by: vemkiran
6 Replies

10. UNIX for Advanced & Expert Users

sed: -e expression #1, char 0: no previous regular expression

Hello All, I'm trying to extract the lines between two consecutive elements of an array from a file. My array looks like: problem_arr=(PRS111 PRS213 PRS234) j=0 while } ] do k=`expr $j + 1` sed -n "/${problem_arr}/,/${problem_arr}/p" problemid.txt ---some operation goes... (11 Replies)
Discussion started by: InduInduIndu
11 Replies
dos2unix(1)							    2010-04-03							       dos2unix(1)

NAME
dos2unix - DOS/MAC to UNIX and vice versa text file format converter SYNOPSIS
dos2unix [options] [-c CONVMODE] [-o FILE ...] [-n INFILE OUTFILE ...] unix2dos [options] [-c CONVMODE] [-o FILE ...] [-n INFILE OUTFILE ...] DESCRIPTION
The Dos2unix package includes utilities "dos2unix" and "unix2dos" to convert plain text files in DOS or MAC format to UNIX format and vice versa. Binary files and non-regular files, such as soft links, are automatically skipped, unless conversion is forced. Dos2unix has a few conversion modes similar to dos2unix under SunOS/Solaris. In DOS/Windows text files line endings exist out of a combination of two characters: a Carriage Return (CR) followed by a Line Feed (LF). In Unix text files line endings exists out of a single Newline character which is equal to a DOS Line Feed (LF) character. In Mac text files, prior to Mac OS X, line endings exist out of a single Carriage Return character. Mac OS X is Unix based and has the same line endings as Unix. OPTIONS
-c, --convmode CONVMODE Set conversion mode. Where CONVMODE is one of: ascii, 7bit, iso, mac with ascii being the default. -f, --force Force conversion of all files. Also binary files. -h, --help Display online help. -k, --keepdate Keep the date stamp of output file same as input file. -L, --license Display software license. -l, --newline Add additional newline. dos2unix: Only DOS line endings are changed to two Unix line endings. In Mac mode only Mac line endings are changed to two Unix line endings. unix2dos: Only Unix line endings are changed to two DOS line endings. In Mac mode Unix line endings are changed to two Mac line endings. -n, --newfile INFILE OUTFILE ... New file mode. Convert the infile and write output to outfile. File names must be given in pairs and wildcard names should NOT be used or you WILL lose your files. -o, --oldfile FILE ... Old file mode. Convert the file and write output to it. The program default to run in this mode. Wildcard names may be used. -q, --quiet Quiet mode. Suppress all warning and messages. -V, --version Display version information. CONVERSION MODES
Conversion modes ascii, 7bit, and iso are similar to those of dos2unix/unix2dos under SunOS/Solaris. ascii dos2unix: In this mode DOS line endings are converted to Unix line endings. Unix and Mac line endings are not changed. unix2dos: In this mode Unix line endings are converted to DOS line endings. DOS and Mac line endings are not changed. Although the name of this mode is ASCII, which is a 7 bit standard, the actual mode is 8 bit. mac dos2unix: In this mode Mac line endings are converted to Unix line endings. DOS and Unix line endigs are not changed. You can also use the command "mac2unix" to run dos2unix in Mac mode. unix2dos: In this mode Unix line endings are converted to Mac line endings. DOS and Mac line endigs are not changed. You can also use the command "unix2mac" to run unix2dos in Mac mode. 7bit In this mode DOS line endings are converted to Unix line endings or vice versa. All 8 bit non-ASCII characters (with values from 128 to 255) are converted to a space. iso In this mode DOS line endings are converted to Unix line endings or vice versa. Characters are converted between the DOS character set (code page) CP437 and ISO character set ISO-8859-1 on Unix. CP437 characters without ISO-8859-1 equivalent, for which conversion is not possible, are converted to a dot. The same counts for ISO-8859-1 characters without CP437 counterpart. CP437 is mainly used in the USA. In Western Europe CP850 is more standard. Another option to convert text files between different encodings is to use dos2unix in combination with iconv(1). Iconv can convert between a long list of character encodings. Some examples: Convert from DOS DOSLatinUS to Unix Latin-1 iconv -f CP437 -t ISO-8859-1 in.txt | dos2unix > out.txt Convert from DOS DOSLatin1 to Unix Latin-1 iconv -f CP850 -t ISO-8859-1 in.txt | dos2unix > out.txt Convert from Windows WinLatin1 to Unix Latin-1 iconv -f CP1252 -t ISO-8859-1 in.txt | dos2unix > out.txt Convert from Windows WinLatin1 to Unix UTF-8 (Unicode) iconv -f CP1252 -t UTF-8 in.txt | dos2unix > out.txt Convert from Windows UTF-16 (Unicode) to Unix UTF-8 (Unicode) iconv -f UTF-16 -t UTF-8 in.txt | dos2unix > out.txt Convert from Unix Latin-1 to DOS DOSLatinUS unix2dos < in.txt | iconv -f ISO-8859-1 -t CP437 > out.txt Convert from Unix Latin-1 to DOS DOSLatin1 unix2dos < in.txt | iconv -f ISO-8859-1 -t CP850 > out.txt Convert from Unix Latin-1 to Windows WinLatin1 unix2dos < in.txt | iconv -f ISO-8859-1 -t CP1252 > out.txt Convert from Unix UTF-8 (Unicode) to Windows WinLatin1 unix2dos < in.txt | iconv -f UTF-8 -t CP1252 in.txt > out.txt Convert from Unix UTF-8 (Unicode) to Windows UTF-16 (Unicode) unix2dos < in.txt | iconv -f UTF-8 -t UTF-16 > out.txt See also <http://czyborra.com/charsets/codepages.html> and <http://czyborra.com/charsets/iso8859.html>. UNICODE
Unicode files can be encoded in different encodings. On Unix/Linux Unicode files are mostly encoded in UTF-8 encoding. UTF-8 is ASCII compatible. UTF-8 files can be in DOS, Unix or Mac format. It is safe to run dos2unix/unix2dos on UTF-8 encoded files. On Windows mostly UTF-16 encoding is used for Unicode files. Dos2unix/unix2dos should not be run on UTF-16 files. UTF-16 files are automatically skipped, because it are binary files. EXAMPLES
Get input from stdin and write output to stdout. dos2unix dos2unix -l -c mac Convert and replace a.txt. Convert and replace b.txt. dos2unix a.txt b.txt dos2unix -o a.txt b.txt Convert and replace a.txt in ascii conversion mode. dos2unix a.txt Convert and replace a.txt in ascii conversion mode. Convert and replace b.txt in 7bit conversion mode. dos2unix a.txt -c 7bit b.txt dos2unix -c ascii a.txt -c 7bit b.txt Convert a.txt from Mac to Unix format. dos2unix -c mac a.txt mac2unix a.txt Convert a.txt from Unix to Mac format. unix2dos -c mac a.txt unix2mac a.txt Convert and replace a.txt while keeping original date stamp. dos2unix -k a.txt dos2unix -k -o a.txt Convert a.txt and write to e.txt. dos2unix -n a.txt e.txt Convert a.txt and write to e.txt, keep date stamp of e.txt same as a.txt. dos2unix -k -n a.txt e.txt Convert and replace a.txt. Convert b.txt and write to e.txt. dos2unix a.txt -n b.txt e.txt dos2unix -o a.txt -n b.txt e.txt Convert c.txt and write to e.txt. Convert and replace a.txt. Convert and replace b.txt. Convert d.txt and write to f.txt. dos2unix -n c.txt e.txt -o a.txt b.txt -n d.txt f.txt LOCALIZATION
LANG The primary language is selected with the environment variable LANG. The LANG variable consists out of several parts. The first part is in small letters the language code. The second is optional and is the country code in capital letters, preceded with an underscore. There is also an optional third part: character encoding, preceded with a dot. A few examples for POSIX standard type shells: export LANG=nl Dutch export LANG=nl_NL Dutch, The Netherlands export LANG=nl_BE Dutch, Belgium export LANG=es_ES Spanish, Spain export LANG=es_MX Spanish, Mexico export LANG=en_US.iso88591 English, USA, Latin-1 encoding export LANG=en_GB.UTF-8 English, UK, UTF-8 encoding For a complete list of language and country codes see the gettext manual: <http://www.gnu.org/software/gettext/manual/gettext.html#Language-Codes> On Unix systems you can use to command locale(1) to get locale specific information. LANGUAGE With the LANGUAGE environment variable you can specify a priority list of languages, separated by colons. Dos2unix gives preference to LANGUAGE over LANG. For instance, first Dutch and then German: "LANGUAGE=nl:de". You have to first enable localization, by setting LANG (or LC_ALL) to a value other than "C", before you can use a language priority list through the LANGUAGE variable. See also the gettext manual: <http://www.gnu.org/software/gettext/manual/gettext.html#The-LANGUAGE-variable> For Esperanto there is a special language file in x-method format. X-method can be used on systems that don't support Latin-3 or Unicode character encoding. Make LANGUAGE equal to "eo-x:eo". If you select a language which is not available you will get the standard English messages. DOS2UNIX_LOCALEDIR With the environment variable DOS2UNIX_LOCALEDIR the LOCALEDIR set during compilation can be overruled. LOCALEDIR is used to find the language files. The GNU default value is "/usr/local/share/locale". Option "-V" will display the LOCALEDIR that is used. Example (windows cmd): set DOS2UNIX_LOCALEDIR=c:/my_prefix/share/locale AUTHORS
Benjamin Lin - <blin@socs.uts.edu.au> Bernd Johannes Wuebben (mac2unix mode) - <wuebben@kde.org> Erwin Waterlander - <waterlan@xs4all.nl> Project page: <http://www.xs4all.nl/~waterlan/dos2unix.html> SourceForge page: <http://sourceforge.net/projects/dos2unix/> Freshmeat: <http://freshmeat.net/projects/dos2unix> SEE ALSO
iconv(1) dos2unix 2010-03-23 dos2unix(1)
All times are GMT -4. The time now is 02:01 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy