diff options
Diffstat (limited to 'JLanguageTool/src/resource/de/format_fsa.awk')
-rw-r--r-- | JLanguageTool/src/resource/de/format_fsa.awk | 24 |
1 files changed, 24 insertions, 0 deletions
diff --git a/JLanguageTool/src/resource/de/format_fsa.awk b/JLanguageTool/src/resource/de/format_fsa.awk new file mode 100644 index 0000000..4bea30c --- /dev/null +++ b/JLanguageTool/src/resource/de/format_fsa.awk @@ -0,0 +1,24 @@ +# called from prepare_fsa_format.sh +BEGIN { + FS=" " +} +#getting lemma +#format: +#form +#lemma POS_tags base lemma (optional) +#@ +# +!/\*/{if (NF==1 && $1!="@") fullform = $1 + +if (NF>1) { + printf fullform"\t"$1"\t" + gsub(",","+") #DAT,AKK + tags="" + for (i=2;i<=(NF-1);i++) + tags=tags $i":" + if (toupper($NF)==$NF && $NF!~/[0-9]/) tags = tags $NF + #trailing : + gsub(/:$/,"", tags) + printf tags "\n" + } +} |