Skip to content

Commit

Permalink
lost bits
Browse files Browse the repository at this point in the history
  • Loading branch information
@flammie
flammie committed Mar 27, 2024
1 parent 6aecb4c commit d7f1f79
Showing 1 changed file with 182 additions and 3 deletions.
185 changes: 182 additions & 3 deletions src/fst/Makefile.am
View file
Original file line number Diff line number Diff line change
Expand Up @@ -40,9 +40,6 @@ CUSTOM_FSTS=
#### BEGIN: Add local processing instructions BELOW this line ####
##################################################################

########################################################
#### Add language-specific transducer targets here: ####

#### Xerox transducers:
if CAN_XFST
GT_ANALYSERS+=
Expand Down Expand Up @@ -79,8 +76,190 @@ endif # CAN_FOMA
#################################################
#### Add language-specific build rules here: ####


### Split multichar letters early, to avoid repetitive code. Multichar ###
### letters must be split on both sides, and then the alphabet pruned, ###
### for hfst-tokenise to work without issues. ###

# Hfst - add weights to compounds if using tropical-semiring fst format:
if WITH_OFST_TROPICAL
.generated/generator-raw-gt-desc.hfst: .generated/generator-raw-gt-desc.tmp.hfst \
orthography/split-composed-chars.compose.hfst
$(AM_V_XFST_TOOL)$(PRINTF) "read regex \
@\"orthography/split-composed-chars.compose.hfst\".i \
.o. @\"$<\" \
.o. @\"orthography/split-composed-chars.compose.hfst\" \
;\n\
save stack $@.tmp\n\
quit\n" | $(XFST_TOOL)
$(AM_V_REWEIGHT)$(HFST_PRUNE_ALPHABET) -i $@.tmp \
$(HFST_REWEIGHT) $(MORE_VERBOSITY) $(HFST_FLAGS) \
-S '+Cmp' -a 10 --arcs-only \
> $@
$(AM_V_at)rm -f $@.tmp

else

.generated/generator-raw-gt-desc.hfst: .generated/generator-raw-gt-desc.tmp.hfst \
orthography/split-composed-chars.compose.hfst
$(AM_V_XFST_TOOL)$(PRINTF) "read regex \
@\"orthography/split-composed-chars.compose.hfst\".i \
.o. @\"$<\" \
.o. @\"orthography/split-composed-chars.compose.hfst\" \
;\n\
save stack $@.tmp\n\
quit\n" | $(XFST_TOOL)
$(AM_V_HPRUNE)$(HFST_PRUNE_ALPHABET) -i $@.tmp -o $@
$(AM_V_at)rm -f $@.tmp

endif

.generated/analyser-raw-gt-desc.%: .generated/analyser-raw-gt-desc.tmp.% \
orthography/split-composed-chars.compose.%
$(AM_V_XFST_TOOL)$(PRINTF) "read regex \
@\"orthography/split-composed-chars.compose.$*\".i \
.o. @\"$<\" \
.o. @\"orthography/split-composed-chars.compose.$*\" \
;\n\
save stack $@\n\
quit\n" | $(XFST_TOOL)


# Hfst:
.generated/analyser-%.hfst: .generated/analyser-%.tmp.hfst \
filters/remove-combining-caron-below.hfst \
filters/remove-modifier-letter-grave-accent.hfst \
filters/remove-combining-circumflex-accent-below.hfst
$(AM_V_RGX2FST)$(PRINTF) " \
@\"$<\" \
.o. @\"filters/remove-combining-circumflex-accent-below.hfst\" \
.o. @\"filters/remove-combining-caron-below.hfst\" \
.o. @\"filters/remove-modifier-letter-grave-accent.hfst\" \
;" \
| $(HFST_REGEXP2FST) $(HFST_FLAGS) $(HFST_FORMAT) \
-S --xerox-composition=ON \
| $(HFST_INVERT) \
> $@

# We need to specify -gt- to avoid applying the filters to the raw fst:
.generated/generator-gt-%.hfst: .generated/generator-gt-%.tmp.hfst \
filters/remove-combining-caron-below.hfst \
filters/remove-modifier-letter-grave-accent.hfst \
filters/remove-combining-circumflex-accent-below.hfst
$(AM_V_RGX2FST)$(PRINTF) " \
@\"$<\" \
.o. @\"filters/remove-combining-circumflex-accent-below.hfst\" \
.o. @\"filters/remove-combining-caron-below.hfst\" \
.o. @\"filters/remove-modifier-letter-grave-accent.hfst\" \
;" \
| $(HFST_REGEXP2FST) $(HFST_FLAGS) $(HFST_FORMAT) \
-S --xerox-composition=ON \
> $@

# Xerox:
# We need to specify -gt- to avoid applying the filters to the raw fst:
.generated/analyser-gt-%.xfst: .generated/analyser-gt-%.tmp.xfst \
filters/remove-combining-caron-below.xfst \
filters/remove-modifier-letter-grave-accent.xfst \
filters/remove-combining-circumflex-accent-below.xfst
$(AM_V_XFST)$(PRINTF) "read regex \
@\"$<\" \
.o. @\"filters/remove-combining-circumflex-accent-below.xfst\" \
.o. @\"filters/remove-combining-caron-below.xfst\" \
.o. @\"filters/remove-modifier-letter-grave-accent.xfst\" \
;\n\
save stack $@\n\
quit\n" | $(XFST) $(VERBOSITY)

.generated/generator-%.xfst: .generated/generator-%.tmp.xfst \
filters/remove-combining-caron-below.xfst \
filters/remove-modifier-letter-grave-accent.xfst \
filters/remove-combining-circumflex-accent-below.xfst
$(AM_V_XFST)$(PRINTF) "read regex \
@\"$<\" \
.o. @\"filters/remove-combining-circumflex-accent-below.xfst\" \
.o. @\"filters/remove-combining-caron-below.xfst\" \
.o. @\"filters/remove-modifier-letter-grave-accent.xfst\" \
;\n\
invert net\n\
save stack $@\n\
quit\n" | $(XFST) $(VERBOSITY)

# Special case for the disamb analyser, since it follows the same filename
# pattern as the raw fst:
.generated/analyser-disamb-gt-%.xfst: .generated/analyser-disamb-gt-%.tmp.xfst \
filters/remove-combining-caron-below.xfst \
filters/remove-modifier-letter-grave-accent.xfst \
filters/remove-combining-circumflex-accent-below.xfst
$(AM_V_XFST)$(PRINTF) "read regex \
@\"$<\" \
.o. @\"filters/remove-combining-circumflex-accent-below.xfst\" \
.o. @\"filters/remove-combining-caron-below.xfst\" \
.o. @\"filters/remove-modifier-letter-grave-accent.xfst\" \
;\n\
save stack $@\n\
quit\n" | $(XFST) $(VERBOSITY)

# Foma, for completeness:
# We need to specify -gt- to avoid applying the filters to the raw fst:
.generated/analyser-gt-%.foma: .generated/analyser-gt-%.tmp.foma \
filters/remove-combining-caron-below.foma \
filters/remove-modifier-letter-grave-accent.foma \
filters/remove-combining-circumflex-accent-below.foma
$(AM_V_XFST)$(PRINTF) "read regex \
@\"$<\" \
.o. @\"filters/remove-combining-circumflex-accent-below.foma\" \
.o. @\"filters/remove-combining-caron-below.foma\" \
.o. @\"filters/remove-modifier-letter-grave-accent.foma\" \
;\n\
save stack $@\n\
quit\n" | $(XFST) $(VERBOSITY)

.generated/generator-%.foma: .generated/generator-%.tmp.foma \
filters/remove-combining-caron-below.foma \
filters/remove-modifier-letter-grave-accent.foma \
filters/remove-combining-circumflex-accent-below.foma
$(AM_V_XFST)$(PRINTF) "read regex \
@\"$<\" \
.o. @\"filters/remove-combining-circumflex-accent-below.foma\" \
.o. @\"filters/remove-combining-caron-below.foma\" \
.o. @\"filters/remove-modifier-letter-grave-accent.foma\" \
;\n\
invert net\n\
save stack $@\n\
quit\n" | $(XFST) $(VERBOSITY)

# Special case for the disamb analyser, since it follows the same filename
# pattern as the raw fst:
.generated/analyser-disamb-gt-%.foma: .generated/analyser-disamb-gt-%.tmp.foma \
filters/remove-combining-caron-below.foma \
filters/remove-modifier-letter-grave-accent.foma \
filters/remove-combining-circumflex-accent-below.foma
$(AM_V_XFST)$(PRINTF) "read regex \
@\"$<\" \
.o. @\"filters/remove-combining-circumflex-accent-below.foma\" \
.o. @\"filters/remove-combining-caron-below.foma\" \
.o. @\"filters/remove-modifier-letter-grave-accent.foma\" \
;\n\
save stack $@\n\
quit\n" | $(XFST) $(VERBOSITY)

# Do NOT apply the accent removal filters to the normative
# dictionary generator (both hfst and xfst/foma):
.generated/generator-dict-gt-norm.hfst: .generated/generator-dict-gt-norm.tmp.hfst
$(AM_V_CP)cp -f $< $@

.generated/generator-dict-gt-norm.%: .generated/generator-dict-gt-norm.tmp.%
$(AM_V_XFST)$(PRINTF) "\
load stack $<\n\
invert net\n\
save stack $@\n\
quit\n" | $(XFST) $(VERBOSITY)


##################################################################
#### END: Add local processing instructions ABOVE this line ######
##################################################################


include $(top_srcdir)/../giella-core/am-shared/src-fst-dir-include.am

0 comments on commit d7f1f79

Please sign in to comment.