Datasets:
cjvt
/

Tasks:
Other
License:
Matej Klemen commited on
Commit
aab5879
1 Parent(s): fc2cf60

Modify script to enable loading all languages supported by Parlamint3

Browse files
Files changed (1) hide show
  1. parlaMintSI.py → ParlaMint3.py +47 -33
parlaMintSI.py → ParlaMint3.py RENAMED
@@ -14,7 +14,6 @@
14
  # limitations under the License.
15
 
16
  import csv
17
- import json
18
  import os
19
 
20
  import datasets
@@ -41,8 +40,7 @@ The corpora are also marked to the subcorpus they belong to ("reference", until
41
  The corpora are encoded according to the Parla-CLARIN TEI recommendation (https://clarin-eric.github.io/parla-clarin/), but have been encoded against the compatible,
42
  but much stricter ParlaMint encoding guidelines (https://clarin-eric.github.io/ParlaMint/) and schemas (included in this distribution).
43
  This entry contains the ParlaMint TEI-encoded corpora with the derived plain text versions of the corpora along with TSV metadata of the speeches.
44
- Also included is the 3.0 release of the data and scripts available at the GitHub repository of the ParlaMint project.
45
- This dataset contains only Slovenian parliamentary debates.
46
  """
47
 
48
 
@@ -50,15 +48,22 @@ _HOMEPAGE = "http://hdl.handle.net/11356/1486"
50
 
51
  _LICENSE = "Creative Commons - Attribution 4.0 International (CC BY 4.0)"
52
 
 
 
 
53
  _URLS = {
54
- "parlamint": "https://www.clarin.si/repository/xmlui/bitstream/handle/11356/1486/ParlaMint-SI.tgz?sequence=24&isAllowed=y",
 
55
  }
56
 
57
 
58
- class ParlaMintSI(datasets.GeneratorBasedBuilder):
59
  """This dataset contains transcriptions of Slovenian parliamentary debates and relevant metadata."""
60
 
61
- VERSION = datasets.Version("1.1.0")
 
 
 
62
 
63
  def _info(self):
64
  features = datasets.Features(
@@ -69,7 +74,7 @@ class ParlaMintSI(datasets.GeneratorBasedBuilder):
69
  "Body": datasets.Value("string"),
70
  "Term": datasets.Value("string"),
71
  "Session": datasets.Value("string"),
72
- "Meeting": datasets.Value("int32"),
73
  "Sitting": datasets.Value("string"),
74
  "Agenda": datasets.Value("string"),
75
  "Subcorpus": datasets.Value("string"),
@@ -95,35 +100,44 @@ class ParlaMintSI(datasets.GeneratorBasedBuilder):
95
  )
96
 
97
  def _split_generators(self, dl_manager):
98
- urls = _URLS["parlamint"]
 
99
  download_path = dl_manager.download_and_extract(urls)
100
  return [
101
  datasets.SplitGenerator(
102
  name=datasets.Split.TRAIN,
103
- gen_kwargs={
104
- "filepath": download_path,
105
- },
106
- ),
107
  ]
108
 
109
- def _generate_examples(self, filepath):
110
- filepath = os.path.join(filepath, "ParlaMint-SI.txt")
111
-
112
- for year_dir in os.listdir(filepath):
113
- year_path = os.path.join(filepath, year_dir)
114
- if os.path.isdir(year_path):
115
- tsv_files = [f for f in os.listdir(year_path) if f.endswith(".tsv")]
116
- for tsv_file in tsv_files:
117
- tsv_path = os.path.join(year_path, tsv_file)
118
- txt_path = os.path.join(year_path, tsv_file.replace("-meta.tsv", ".txt"))
119
-
120
- with open(tsv_path, "r", encoding="utf-8") as tsv, open(txt_path, "r", encoding="utf-8") as txt:
121
- tsv_reader = csv.DictReader(tsv, delimiter="\t")
122
- txt_content = txt.readlines()
123
-
124
- for row in tsv_reader:
125
- id_ = row.get("ID", "")
126
- text = next((line.split("\t")[1] for line in txt_content if line.startswith(id_)), "")
127
- example = {key: row.get(key, "") for key in row}
128
- example["text"] = text
129
- yield id_, example
 
 
 
 
 
 
 
 
 
 
 
14
  # limitations under the License.
15
 
16
  import csv
 
17
  import os
18
 
19
  import datasets
 
40
  The corpora are encoded according to the Parla-CLARIN TEI recommendation (https://clarin-eric.github.io/parla-clarin/), but have been encoded against the compatible,
41
  but much stricter ParlaMint encoding guidelines (https://clarin-eric.github.io/ParlaMint/) and schemas (included in this distribution).
42
  This entry contains the ParlaMint TEI-encoded corpora with the derived plain text versions of the corpora along with TSV metadata of the speeches.
43
+ Also included is the 3.0 release of the data and scripts available at the GitHub repository of the ParlaMint project.
 
44
  """
45
 
46
 
 
48
 
49
  _LICENSE = "Creative Commons - Attribution 4.0 International (CC BY 4.0)"
50
 
51
+ SUPPORTED_LANGS = ["at", "ba", "be", "bg", "cz", "dk", "ee", "es-ct", "es-ga", "fr", "gb", "gr", "hr", "hu", "is",
52
+ "it", "lv", "nl", "no", "pl", "pt", "rs", "se", "si", "tr", "ua"]
53
+
54
  _URLS = {
55
+ lang: f"https://www.clarin.si/repository/xmlui/bitstream/handle/11356/1486/ParlaMint-{lang.upper()}.tgz"
56
+ for lang in SUPPORTED_LANGS
57
  }
58
 
59
 
60
+ class ParlaMint3(datasets.GeneratorBasedBuilder):
61
  """This dataset contains transcriptions of Slovenian parliamentary debates and relevant metadata."""
62
 
63
+ BUILDER_CONFIGS = [
64
+ datasets.BuilderConfig(name=lang, version=datasets.Version("1.2.0"), description=f"{lang} parliamentary corpus")
65
+ for lang in SUPPORTED_LANGS
66
+ ]
67
 
68
  def _info(self):
69
  features = datasets.Features(
 
74
  "Body": datasets.Value("string"),
75
  "Term": datasets.Value("string"),
76
  "Session": datasets.Value("string"),
77
+ "Meeting": datasets.Value("string"),
78
  "Sitting": datasets.Value("string"),
79
  "Agenda": datasets.Value("string"),
80
  "Subcorpus": datasets.Value("string"),
 
100
  )
101
 
102
  def _split_generators(self, dl_manager):
103
+ urls = _URLS[self.config.name]
104
+
105
  download_path = dl_manager.download_and_extract(urls)
106
  return [
107
  datasets.SplitGenerator(
108
  name=datasets.Split.TRAIN,
109
+ gen_kwargs={"data_dir": download_path}
110
+ )
 
 
111
  ]
112
 
113
+ def _generate_examples(self, data_dir):
114
+ data_dir = os.path.join(data_dir, f"ParlaMint-{self.config.name.upper()}.txt")
115
+
116
+ years = [curr_dir for curr_dir in os.listdir(data_dir) if os.path.isdir(os.path.join(data_dir, curr_dir))]
117
+ years = sorted(years, key=lambda _yr: int(_yr))
118
+
119
+ for year_dir in years:
120
+ # Metadata inside tab-separated files
121
+ tsv_files = sorted([f for f in os.listdir(os.path.join(data_dir, year_dir)) if f.endswith(".tsv")])
122
+
123
+ for fname in tsv_files:
124
+ tsv_path = os.path.join(data_dir, year_dir, fname)
125
+ # Text data inside txt files
126
+ txt_path = os.path.join(data_dir, year_dir, fname.replace("-meta.tsv", ".txt"))
127
+
128
+ with open(tsv_path, "r", encoding="utf-8") as f_tsv, \
129
+ open(txt_path, "r", encoding="utf-8") as f_txt:
130
+ tsv_reader = csv.DictReader(f_tsv, delimiter="\t")
131
+
132
+ txt_content = {} # ID of utterance -> text of utterance
133
+ for _line in f_txt:
134
+ _parts = _line.strip().split("\t")
135
+ txt_content[_parts[0]] = _parts[1]
136
+
137
+ for row in tsv_reader:
138
+ _id = row["ID"]
139
+ text = txt_content[_id]
140
+ example = {key: row.get(key, "") for key in row}
141
+ example["text"] = text
142
+
143
+ yield _id, example