Hi, Thanks for your reply to my last mail. i used the MAP and PHONE table in my affix file and got desired result.I also put my affix and dictionary file to replace default dic file and affix file in OpenOffice.org writer and it works but is suggesting the words which i got in command line before using the MAP and PHONE table .The affix file that i used for the command and the i used for OpenOffice are the same. i don't know where i made mistake. Is possible to use prefixes in affix file as a valid word in the dictionary file. For the displaying of all the valid words , i copied the wordforms file into the tools directory in the hunspell packacge in cygwin and execute the command as below : $ chmod a+x wordforms $ ./wordforms en_US.aff en_US.dic word >output.txt
when i checked the file it is an empty file. please what can i do again. Best regard jeje --- On Mon, 10/6/08, Németh László <[EMAIL PROTECTED]> wrote: From: Németh László <[EMAIL PROTECTED]> Subject: Re: [lingu-dev] ASSISTANCE IN USING HUNSPELL To: [email protected], [EMAIL PROTECTED] Date: Monday, October 6, 2008, 1:27 PM Hi, (1) This shell script generates all (except double suffixed) forms of a word: http://downloads.sourceforge.net/hunspell/wordforms Usage (on Linux or in Cygwin environment on Windows): $ chmod a+x wordforms $ ./wordforms en_US.aff en_US.dic word reword words wordings word worded wording rewords rewordings reword reworded rewording (2) Use MAP and PHONE features in the affix file. MAP suggestion search checks all combinations of the MAP character equivalents in the word. A possible MAP definition for your language: MAP 11 MAP àaá MAP óoòọ̀ọọ́ MAP ìì MAP ṣs MAP ẹ̀ẹ́ẹèée MAP ùúu MAP SṢ MAP ÀÁA MAP ÒÓOỌ̀Ọ́Ọ MAP ÌÍI MAP ÈÉE PHONE suggestion search checks all dictionary words transformed by the PHONE table and using n-gram similarity. Suggested PHONE table for your language (bracket expression hasn't been supported in UTF-8 yet): PHONE 29 PHONE à a PHONE á a PHONE ó o PHONE ò o PHONE ọ̀ o PHONE ọ o PHONE ọ́ o PHONE ì i PHONE ì i PHONE ṣ s PHONE ẹ̀ e PHONE ẹ́ e PHONE ẹ e PHONE è e PHONE é e PHONE ù u PHONE ú u PHONE Ṣ S PHONE À A PHONE Á A PHONE Ò O PHONE Ó O PHONE Ọ̀ O PHONE Ọ́ O PHONE Ọ O PHONE Ì I PHONE Í I PHONE È E PHONE É E > How can modify base dictionary of the OpenOffice.Org with my affix rule and my dictionary. 1. For a quick test, replace default dictionary (for example en_US.aff and en_US.dic) of your OpenOffice.org installation with your renamed dictionary files. 2. For the upcoming OpenOffice.org 3, you need a dictionary extension, see http://wiki.services.openoffice.org/wiki/Extensions_development Maybe your language hasn't been in the language list box yet. In this case, make a new issue for this task (see http://www.openoffice.org/issues/show_bug.cgi?id=80114 for example). 3. With own OpenOffice.org builds you can add default dictionary package to your installation set: http://wiki.services.openoffice.org/wiki/Spellchecker_Integration_into_Installation_Set 4. For a native-language OpenOffice.org, joining the NLC: http://native-lang.openoffice.org/ Best regards, László 2008/9/22 Sunday Bolaji <[EMAIL PROTECTED]>: > Hi, > please what can i do to able to do these things using hunspell. > > (1) display all the valid words both the words in dictionary file and the the words generate using affix rule. > (2) suggest word starting with tone mark, if i type the word but without tone mark as shown below. > " > aiku > & aiku 7 0: akun, asu, awun, asun,àìkù, àìkú,àìkun " > when i typed the word "aiku" i got what i stated below: > " & aiku 4 0: akun, aṣu, awun, asun" > > (3)suggest a word with under tone character if type the same word without under tone as below also: > " > se > & se 5 0: sè, sé, ṣe ,sé,s è,ṣè"̣ > > (4)sugget a word with both tone mark and under tone character if i type the same word but without underton and tone mark as shown in the example below. > > " eje 5 :èjé,ejè,ẹjẹ,ẹ̀jẹ̀,ẹjẹ́" > > Below are the sample of my REP table and TRY characters. > ̣ > TRY tmnkwlbàaáóoòprọ̀ọọ́ììfdyṣsgẹ̀ẹ́ẹèéeùúuTMNṢLBÀÁAÒÓOPRỌ̀Ọ́ỌÌÍIGÈÉE > > REP 38 > REP a à > REP à á > REP a á > REP á à > REP e è > REP è é > REP e é > REP é è > REP ẹ ẹ̀ > REP ẹ̀ ẹ́ > REP ẹ ẹ́ > REP ẹ́ ẹ̀ > REP i ì > REP ì í > REP i í > REP í ì > REP e ẹ > REP è ẹ̀ > REP é ẹ́ > REP o ò > REP ò ó > REP o ó > REP ó ò > REP ọ ọ̀ > REP ọ̀ ọ́ > REP ọ ọ́ > REP ọ́ ọ̀ > REP o ọ > REP ò ọ̀ > REP ó ọ́ > REP s ç > REP ai àì > REP e ẹ̀ > REP e ẹ́ > REP o ọ̀ > REP o ọ́ > REP u ù > REP u ú > > > Best regards > > > > > > > > > > > --- On Tue, 8/26/08, Németh László <[EMAIL PROTECTED]> wrote: > From: Németh László <[EMAIL PROTECTED]> > Subject: Re: [lingu-dev] ASSISTANCE IN USING HUNSPELL > To: [email protected], [EMAIL PROTECTED] > Date: Tuesday, August 26, 2008, 5:28 AM > > Hi Jeje, > >> Please can i get manual that explains better all the options explain in > tests folder of >> hunspell distribution. some areas are not clear as explain in manual4. > Also how can i >> test for words that contain unicode character in a command line. is there > any >> graphical user interface of hunspell to test for this word. > > Please, specify the problematic features. > > I believe, there is no problem with Unicode encoding under Linux. On > Windows, see my previous letter about using input and output redirections. > Abiword, Firefox 3 and OpenOffice.org have Unicode support and contain > Hunspell. You can modify their base dictionaries with your affix rules and > words. Eleonóra Goldman has Perl/tk and Java based Hunspell clients > (Ptkspell and Huncheck) here: http://tkltrans.sourceforge.net/spell.htm. > > 2008/8/22 Sunday Bolaji <[EMAIL PROTECTED]> > >> Hi Laszlo, >> Please is there any way to use hunspell to create the >> following affix rules : >> (1) that will take first character of word (consonant) and combine with >> vowel 'i' and use to prefix the word. as shown in the example > below. >> >> >> wi + wa------wíwá >> >> Li + lo -------lílo >> (2) that will combine the first character of word with 'L' and use > it to >> prefix the word as shown also below. >> >> >> al >> + ata-----aláta >> >> el >> + edé------elédé >> >> >> only one rule that will this for all word in each case. > > > Unfortunately, Hunspell hasn't supported the planned infix rules yet, so > instead of the future > > PFX 0 .í [wl] > PFX 0 .l [ae] > > you need to write > > PFX 0 wí w > PFX 0 lí l > PFX 0 al a > PFX 0 el e > > etc. > > But it needs separated rules for modified diacritics: > PFX a alá a > PFX e elé e > > Best regards, > László > > >> >> >> >> Best regards >> >> >> >> >> >> >> >> >> >> >> >> >> >> --- On Wed, 8/13/08, Németh László <[EMAIL PROTECTED]> wrote: >> From: Németh László <[EMAIL PROTECTED]> >> Subject: Re: [lingu-dev] ASSISTANCE IN USING HUNSPELL >> To: [email protected], [EMAIL PROTECTED] >> Date: Wednesday, August 13, 2008, 3:20 AM >> >> Hi, >> >> PFX B header of your affix file is corrupted: >> >> > PFX B Y >> > 1 >> >> instead of >> >> PFX B Y 1 >> >> Hunspell can help to detect these syntax errors. >> Use --with-warnings compile option for Hunspell warning messages: >> >> cd hunspell-1.2.6 >> make clean >> ./configure --with-warnings >> make >> >> Best regards, >> László >> >> >> 2008/8/13 Sunday Bolaji <[EMAIL PROTECTED]> >> >> > Hi Eleonora, >> > thanks for the the explanation on hunspell. i really >> > appreciate it.But the problem i have now is that hunspell programe >> > suggests only the word in the testing dictionary, it did not generate >> word >> > from affix rule for the language i ' m building the dictionary > for. >> I >> > tested it >> > with English dictionary and affix files, it worked very well.Below > are >> > testing affix and dictionary files. >> > >> > >> > TESTING AFFIX RULE >> > SET ISO8859-1 >> > TRY aeiouAEIOU >> > >> > PFX A Y 1 >> > PFX A 0 wi >> > >> > PFX B Y >> > 1 >> > PFX B 0 li >> > >> > SFX R N 1 >> > SFX R 0 ra >> > >> > SFX Z N 1 >> > SFX Z 0 re >> > >> > PFX D N 1 >> > PFX D 0 e >> > >> > PFX I N 1 >> > PFX I 0 i >> > >> > >> > TESTING DICT >> > 6 >> > wa/ARDI >> > we/AZDI >> > wi/AI >> > wo/ABI >> > wu/AZID >> > lo/BRZI >> > >> > TESTING WORD >> > wa >> > we >> > wi >> > wo >> > wu >> > ewe >> > ewa >> > dwe >> > iwu >> > iwo >> > >> > after i executing this command "example yor_NIGG.aff > yor_NIGG.dic >> > my_file.txt" it gives >> > >> > " " is not correct >> > suggesting we >> > >> > like this for word in dictionary file only. >> > >> > >> > --- On Thu, 8/7/08, ge <[EMAIL PROTECTED]> wrote: >> > From: ge <[EMAIL PROTECTED]> >> > Subject: [lingu-dev] ASSISTANCE IN USING HUNSPELL >> > To: [email protected] >> > Date: Thursday, August 7, 2008, 11:31 AM >> > >> > Hi, >> > >> > Some hints: >> > >> > 1. Check if hunspell.so was built. >> > 2. check if example or example.exe was built. >> > 3. Check if you have >> > en_US.aff and en_US.dic in the tree. >> > if not: download them from OpenOffice >> > 4. Run example or example.exe with proper arguments >> > example /full_path/en_US.aff /full_path/en_US.dic your_file.txt >> > where your_file.txt is a text file for example: >> > tree >> > house >> > treei >> > example should indicate, that treei is bad, rest good. >> > 5. Now you can start to develop yoruba.aff and yoruba.dic and do > tests. >> > >> > Good Luck! >> > >> > regards: eleonora >> > >> > >> > >> > Hi, >> > My name is Abolaji Jeje and i 'm working on project which is >> > spelling checker >> > for my language which is "YORUBA LANGUAGE".I want to use >> hunspell >> > package for this project but i 'm having problem in install the >> > hunspell package.Please kindly assist me. >> > These are the effort i have made in installing the hunspell package > and >> the >> > result i got: >> > >> > CYGWIN ENVIRONMENT >> > After installed the cygwin with >> > all the necessary packages then i ran >> > the following commands: >> > /configure >> > make >> > make check >> > >> > at the end all the activities computer wrote "68 of 86 tests >> fail" >> > if i try to run some executable files computer will write >> "can't >> > open affix and dict files" >> > >> > UBUNTU(LINUX) ENVIROMENT >> > In >> > this environment i ran the command "./configure" and after >> > checking >> > for resources computer wrote "configure error: C++ compiler > cannot >> > create executable: check config.log for details " >> > >> > please kindly assist me to install the hunspell package and how to > use >> to >> > build a spelling checker for my language. >> > >> > Best Regards >> > Jeje Abolaji >> > No 11 oluyole way,Bodija,Ibadan, >> > Oyo state,Nigeria. >> > >> > >> > >> > --------------------------------------------------------------------- >> > To unsubscribe, e-mail: >> > [EMAIL PROTECTED] >> > For additional commands, e-mail: > [EMAIL PROTECTED] >> > >> > >> > >> > >> > >> > >> >> >> >> >> > > > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
