| Marc Kupietz | 4292bc0 | 2026-03-07 16:12:52 +0100 | [diff] [blame] | 1 | SHELL := /bin/bash |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 2 | SRC_DIR ?= I5 |
| Marc Kupietz | e780732 | 2026-06-05 08:55:41 +0200 | [diff] [blame] | 3 | KORAP_PORT ?= 64543 |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 4 | |
| 5 | # Discover all *.i5.xml files in SRC_DIR |
| 6 | I5_FILES := $(wildcard $(SRC_DIR)/*.i5.xml) |
| 7 | BASENAMES := $(patsubst %.i5.xml,%,$(notdir $(I5_FILES))) |
| 8 | |
| Marc Kupietz | 99ff2b0 | 2026-06-05 08:54:37 +0200 | [diff] [blame] | 9 | # Standard TEI P5 support |
| 10 | TEI_DIR ?= TEI |
| 11 | TEI_FILES := $(wildcard $(TEI_DIR)/*.xml) |
| 12 | TEI_ZIP_NAME ?= tei |
| 13 | TEI_FLAGS ?= --auto-textsigle 'TEI/XYZ.00001' # --xmlid-to-textsigle '([A-Z]+)\.(.*)\.([0-9]+)\.*([0-9])@$$1/$$2/$$3$$4' |
| 14 | export TEI_FLAGS |
| 15 | |
| 16 | ifneq ($(TEI_FILES),) |
| 17 | BASENAMES += $(TEI_ZIP_NAME) |
| 18 | endif |
| 19 | |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 20 | BUILD_DIR = build |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 21 | TARGET_DIR ?= ./target |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 22 | MAX_THREADS ?= 8 # $(shell nproc) |
| 23 | MAKE ?= make -j $(shell nproc) |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 24 | # KORAPXMLTOOL_HEAP ?= $(shell echo "$$(($(MAX_THREADS) * 2500))") |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 25 | KORAPXMLTOOL ?= ./bin/korapxmltool |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 26 | KORAPXMLTOOL_MODELS_PATH ?= models |
| Marc Kupietz | 4292bc0 | 2026-03-07 16:12:52 +0100 | [diff] [blame] | 27 | DOCKER_CPU_SHARES ?= # e.g. 512 for lower priority (default Docker value is 1024) |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 28 | |
| 29 | .DELETE_ON_ERROR: |
| 30 | |
| Marc Kupietz | 463da33 | 2026-03-08 13:19:15 +0100 | [diff] [blame] | 31 | .PHONY: all clean test index korap check-src pre-krill krill |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 32 | |
| Marc Kupietz | 5b3ecc0 | 2025-12-15 14:43:09 +0100 | [diff] [blame] | 33 | .PRECIOUS: $(BUILD_DIR)/%.zip $(BUILD_DIR)/%.tree_tagger.zip $(BUILD_DIR)/%.marmot-malt.zip $(BUILD_DIR)/%.spacy.zip $(BUILD_DIR)/%.corenlp.zip $(BUILD_DIR)/%.cmc.zip $(BUILD_DIR)/%.opennlp.zip $(BUILD_DIR)/%.krill.tar %.i5.xml |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 34 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 35 | all: check-src korap |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 36 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 37 | index: check-src $(TARGET_DIR)/index |
| 38 | |
| 39 | check-src: |
| Marc Kupietz | 99ff2b0 | 2026-06-05 08:54:37 +0200 | [diff] [blame] | 40 | @if [ ! -d "$(SRC_DIR)" ] && [ ! -d "$(TEI_DIR)" ]; then \ |
| 41 | echo "Error: Neither SRC_DIR '$(SRC_DIR)' nor TEI_DIR '$(TEI_DIR)' exists."; \ |
| 42 | echo "Please place your .i5.xml files in '$(SRC_DIR)' or your TEI .xml files in '$(TEI_DIR)'."; \ |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 43 | exit 1; \ |
| 44 | fi |
| Marc Kupietz | 99ff2b0 | 2026-06-05 08:54:37 +0200 | [diff] [blame] | 45 | @if [ -z "$$(find "$(SRC_DIR)" -maxdepth 1 -name '*.i5.xml' -print -quit 2>/dev/null)" ] && \ |
| 46 | [ -z "$$(find "$(TEI_DIR)" -maxdepth 1 -name '*.xml' -print -quit 2>/dev/null)" ]; then \ |
| 47 | echo "Error: No .i5.xml files found in '$(SRC_DIR)' and no .xml files found in '$(TEI_DIR)'."; \ |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 48 | exit 1; \ |
| 49 | fi |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 50 | |
| 51 | $(BUILD_DIR)/%.zip: $(SRC_DIR)/%.i5.xml |
| 52 | mkdir -p $(BUILD_DIR) |
| Marc Kupietz | 4292bc0 | 2026-03-07 16:12:52 +0100 | [diff] [blame] | 53 | docker run --rm -i $(if $(DOCKER_CPU_SHARES),--cpu-shares $(DOCKER_CPU_SHARES)) korap/tei2korapxml:latest -l warn -s -tk - < $< > $@ 2> >(tee $(@:.zip=.log) >&2) |
| 54 | # docker run --rm $(if $(DOCKER_CPU_SHARES),--cpu-shares $(DOCKER_CPU_SHARES)) -v $(abspath $<):/input.i5.xml:ro korap/tei2korapxml:latest --progress -l warn -s -tk /input.i5.xml > $@ 2> >(tee $(@:.zip=.log) >&2) |
| 55 | printf "%s\t%s\n" "$$(grep -c '<idsText ' $<)" "$$(unzip -l $@ | grep data.xml | wc -l)" |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 56 | |
| Marc Kupietz | 99ff2b0 | 2026-06-05 08:54:37 +0200 | [diff] [blame] | 57 | $(BUILD_DIR)/$(TEI_ZIP_NAME).zip: $(TEI_FILES) |
| 58 | mkdir -p $(BUILD_DIR) |
| 59 | docker run --rm --entrypoint /bin/sh $(if $(DOCKER_CPU_SHARES),--cpu-shares $(DOCKER_CPU_SHARES)) -v "$(abspath $(TEI_DIR)):/input:ro" korap/tei2korapxml:latest -c 'tei2korapxml -l warn -s -tk '"$$TEI_FLAGS"' /input/*.xml' > $@ 2> >(tee $(@:.zip=.log) >&2) |
| 60 | printf "%s\t%s\n" "$$(cat $^ | grep -c '<teiHeader')" "$$(unzip -l $@ | grep data.xml | wc -l)" |
| 61 | |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 62 | |
| 63 | $(BUILD_DIR)/%.tree_tagger.zip: $(BUILD_DIR)/%.zip bin/korapxmltool |
| Marc Kupietz | 5b3ecc0 | 2025-12-15 14:43:09 +0100 | [diff] [blame] | 64 | $(KORAPXMLTOOL) -j 1 -T treetagger -t zip --force -D $(BUILD_DIR) $< |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 65 | # $(KORAPXMLTOOL) $< | pv | docker run --rm -i korap/conllu2treetagger -l german | conllu2korapxml > $@ |
| 66 | |
| 67 | $(BUILD_DIR)/%.spacy.zip: $(BUILD_DIR)/%.zip bin/korapxmltool |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 68 | $(KORAPXMLTOOL) -P spacy -t zip --force -D $(BUILD_DIR) $< |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 69 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 70 | lib/Krill-Indexer.jar: |
| 71 | mkdir -p lib |
| 72 | curl -sL -o $@ https://github.com/korap/Krill/releases/latest/download/Krill-Indexer.jar |
| Marc Kupietz | 5b3ecc0 | 2025-12-15 14:43:09 +0100 | [diff] [blame] | 73 | |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 74 | bin/korapxmltool: |
| 75 | mkdir -p bin |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 76 | curl -sL -o $@ https://github.com/korap/korapxmltool/releases/latest/download/korapxmltool |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 77 | chmod +x $@ |
| 78 | |
| Marc Kupietz | c523981 | 2026-06-05 08:56:46 +0200 | [diff] [blame^] | 79 | bin/conllu-gender: |
| 80 | mkdir -p bin |
| 81 | curl -sL -o $@ https://github.com/KorAP/conllu-gender/releases/latest/download/conllu-gender |
| 82 | chmod +x $@ |
| 83 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 84 | $(KORAPXMLTOOL_MODELS_PATH)/de.marmot: |
| 85 | mkdir -p $(KORAPXMLTOOL_MODELS_PATH) |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 86 | curl -sL -o $@ https://cistern.cis.lmu.de/marmot/models/CURRENT/spmrl/de.marmot |
| 87 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 88 | $(KORAPXMLTOOL_MODELS_PATH)/german.mco: |
| 89 | mkdir -p $(KORAPXMLTOOL_MODELS_PATH) |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 90 | curl -sL -o $@ https://corpora.ids-mannheim.de/tools/$@ |
| 91 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 92 | $(KORAPXMLTOOL_MODELS_PATH)/dereko_domains_s.classifier: |
| 93 | mkdir -p $(KORAPXMLTOOL_MODELS_PATH) |
| 94 | curl -sL -o $@ https://corpora.ids-mannheim.de/tools/models/$@ |
| 95 | |
| 96 | $(KORAPXMLTOOL_MODELS_PATH)/german-fast.tagger: |
| 97 | mkdir -p $(KORAPXMLTOOL_MODELS_PATH) |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 98 | curl -sL -o $@ https://corpora.ids-mannheim.de/tools/$@ |
| 99 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 100 | $(KORAPXMLTOOL_MODELS_PATH)/germanSR.ser.gz: |
| 101 | mkdir -p $(KORAPXMLTOOL_MODELS_PATH) |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 102 | curl -sL -o $@ https://corpora.ids-mannheim.de/tools/$@ |
| 103 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 104 | $(KORAPXMLTOOL_MODELS_PATH)/de-pos-maxent.bin: |
| 105 | mkdir -p $(KORAPXMLTOOL_MODELS_PATH) |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 106 | curl -sL -o $@ https://corpora.ids-mannheim.de/tools/$@ |
| 107 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 108 | $(BUILD_DIR)/%.marmot-malt.zip: $(BUILD_DIR)/%.zip $(KORAPXMLTOOL_MODELS_PATH)/de.marmot $(KORAPXMLTOOL_MODELS_PATH)/german.mco bin/korapxmltool |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 109 | $(KORAPXMLTOOL) -T marmot:models/de.marmot -P malt:models/german.mco -t zip --force -D $(BUILD_DIR) $< |
| Marc Kupietz | 5b3ecc0 | 2025-12-15 14:43:09 +0100 | [diff] [blame] | 110 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 111 | $(BUILD_DIR)/%.corenlp.zip: $(BUILD_DIR)/%.zip $(KORAPXMLTOOL_MODELS_PATH)/german-fast.tagger $(KORAPXMLTOOL_MODELS_PATH)/germanSR.ser.gz bin/korapxmltool |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 112 | $(KORAPXMLTOOL) -T corenlp -P corenlp -t zip --force -D $(BUILD_DIR) $< |
| 113 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 114 | $(BUILD_DIR)/%.opennlp.zip: $(BUILD_DIR)/%.zip $(KORAPXMLTOOL_MODELS_PATH)/de-pos-maxent.bin bin/korapxmltool |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 115 | $(KORAPXMLTOOL) -T opennlp -t zip --force -D $(BUILD_DIR) $< |
| 116 | |
| Marc Kupietz | 800f9bb | 2025-12-11 17:30:24 +0100 | [diff] [blame] | 117 | $(BUILD_DIR)/%.cmc.zip: $(BUILD_DIR)/%.zip bin/korapxmltool |
| Marc Kupietz | 5b3ecc0 | 2025-12-15 14:43:09 +0100 | [diff] [blame] | 118 | $(KORAPXMLTOOL) -j 1 -A "docker run --rm -i korap/conllu-cmc -s" -l error -F cmc -t zip --force -D $(BUILD_DIR) $< |
| Marc Kupietz | 800f9bb | 2025-12-11 17:30:24 +0100 | [diff] [blame] | 119 | |
| Marc Kupietz | a2876eb | 2026-03-07 22:03:17 +0100 | [diff] [blame] | 120 | $(BUILD_DIR)/%.gender.zip: $(BUILD_DIR)/%.zip bin/conllu-gender |
| Marc Kupietz | 463da33 | 2026-03-08 13:19:15 +0100 | [diff] [blame] | 121 | $(KORAPXMLTOOL) -j 1 -A "bin/conllu-gender -s" -l WARNING -F gender -t zip --force -D $(BUILD_DIR) $< |
| Marc Kupietz | a2876eb | 2026-03-07 22:03:17 +0100 | [diff] [blame] | 122 | |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 123 | # udpipe target removed as requested |
| 124 | # %.ud.zip: %.zip |
| 125 | # $(KORAPXMLTOOL) $< | pv | ./scripts/udpipe2 | conllu2korapxml > $@ |
| 126 | |
| Marc Kupietz | c523981 | 2026-06-05 08:56:46 +0200 | [diff] [blame^] | 127 | # Active annotation layers to run and package into Krill |
| 128 | ANNOTATIONS ?= marmot-malt tree_tagger spacy corenlp opennlp |
| 129 | |
| 130 | KRILL_PREREQS := $(foreach base,$(BASENAMES),$(BUILD_DIR)/$(base).zip $(foreach ann,$(ANNOTATIONS),$(BUILD_DIR)/$(base).$(ann).zip)) |
| Marc Kupietz | 4292bc0 | 2026-03-07 16:12:52 +0100 | [diff] [blame] | 131 | |
| 132 | pre-krill: check-src $(KRILL_PREREQS) |
| 133 | |
| Marc Kupietz | c523981 | 2026-06-05 08:56:46 +0200 | [diff] [blame^] | 134 | $(BUILD_DIR)/%.krill.tar: $(BUILD_DIR)/%.zip $(foreach ann,$(ANNOTATIONS),$(BUILD_DIR)/%.$(ann).zip) |
| Marc Kupietz | 4292bc0 | 2026-03-07 16:12:52 +0100 | [diff] [blame] | 135 | $(KORAPXMLTOOL) --non-word-tokens -f -t krill -D $(BUILD_DIR) $(basename $<)*.zip |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 136 | |
| Marc Kupietz | 463da33 | 2026-03-08 13:19:15 +0100 | [diff] [blame] | 137 | krill: $(foreach base,$(BASENAMES),$(BUILD_DIR)/$(base).krill.tar) |
| 138 | |
| Marc Kupietz | 15b313d | 2025-12-11 12:20:56 +0100 | [diff] [blame] | 139 | $(TARGET_DIR)/index: $(foreach base,$(BASENAMES),$(BUILD_DIR)/$(base).krill.tar) |
| 140 | make lib/Krill-Indexer.jar |
| 141 | touch lib/krill.cfg |
| 142 | mkdir -p $(TARGET_DIR) |
| 143 | java -jar lib/Krill-Indexer.jar -c lib/krill.cfg --progress -i $(subst " ",;,$^) -o $@ |
| 144 | |
| 145 | korap: check-src $(TARGET_DIR)/index |
| Marc Kupietz | e780732 | 2026-06-05 08:55:41 +0200 | [diff] [blame] | 146 | curl -s https://raw.githubusercontent.com/KorAP/KorAP-Docker/master/compose.yaml | sed 's/64543:64543/$(KORAP_PORT):64543/g' | COMPOSE_PROFILES='export' INDEX='$(TARGET_DIR)/index' docker compose -p korap -f - --profile=lite --profile=example up |
| Marc Kupietz | 89a4407 | 2025-12-11 07:14:00 +0100 | [diff] [blame] | 147 | |
| 148 | $(TARGET_DIR)/index.tar.xz: $(TARGET_DIR)/index |
| 149 | tar -I 'xz -T0' -C $(dir $<) -cf $@ $(notdir $<) |
| 150 | |
| 151 | clean: |
| 152 | rm -rf $(BUILD_DIR) $(TARGET_DIR) |