semantic-compressor 2.1__py3-none-any.whl → 2.3__py3-none-any.whl

compressor/resources/nltk_data/tokenizers/punkt_tab/turkish/sent_starters.txt

@@ -0,0 +1,87 @@
+şd
+şimdi
+mahkeme
+ancak
+bilindiği
+arınç
+örneğin
+bunların
+dolayısıyla
+hazine
+anap
+biz
+her
+cumhurbaşkanı
+işte
+hükümet
+diyarbakır
+abd
+artık
+benim
+önümüzdeki
+chp
+bunu
+bence
+akp
+son
+bütün
+irak
+yargıtay
+tüsiad
+ankara
+bu
+iddiaya
+babacan
+tbmm
+bugün
+halbuki
+kanadoğlu
+uzmanlar
+çünkü
+bunun
+burada
+bizim
+buna
+türk
+edinilen
+çiller
+ben
+sezer
+maliye
+başbakan
+yani
+fakat
+diyanet
+imf
+yılmaz
+bana
+ağar
+aslında
+dyp
+bddk
+siz
+gül
+böyle
+eğer
+yeni
+öte
+ab
+sigara
+sonuçta
+bush
+hükümetin
+oysa
+yetkililer
+üstelik
+geçen
+zaten
+böylece
+baykal
+ayrıca
+onlar
+erdoğan
+türkiye
+bunlar
+istanbul
+ama
+ysk

compressor/semantic.py

@@ -14,6 +14,7 @@ from nltk.stem import PorterStemmer
 from nltk.stem import RSLPStemmer
 from collections import Counter
 from model2vec import StaticModel
+import re
 
 tokenizer = RegexTokenizer()
 
@@ -32,6 +33,35 @@ embedding_model = StaticModel.from_pretrained("minishlab/potion-base-2M")
 
 hashing_vectorizer = HashingVectorizer(ngram_range=(1, 6), analyzer='char', n_features=512)
 
+def clean_text(text: str) -> str:
+    # 1) Fix hyphenation at line breaks
+    text = re.sub(r'(\w)-\s*\n\s*(\w)', r'\1\2', text)
+    # 2) Strip stray pipes, bullets, brackets, quotes, unmatched parens
+    text = re.sub(r'[\|\•\[\]\(\)\"“”]', ' ', text)
+    # 3) Remove leading list hyphens
+    text = re.sub(r'(?m)^\s*-\s*', '', text)
+    # 4) Remove hyphens not between letters
+    text = re.sub(r'(?<!\w)-(?!\w)', ' ', text)
+    # 5) Collapse repeated punctuation
+    text = re.sub(r'([!?.,;:]){2,}', r'\1', text)
+    # 6) Normalize whitespace
+    text = re.sub(r'[ \t]+', ' ', text)
+    text = re.sub(r'\n{2,}', '\n', text).strip()
+
+    # 7) Aggressive cleanup if >20% noise, but keep basic punctuation
+    alpha_ratio = sum(c.isalpha() for c in text) / max(len(text), 1)
+    if alpha_ratio < 0.8:
+        text = re.sub(r'[^A-Za-zÀ-ÿ\s\.\,\;\:\?\!]', ' ', text)
+        text = re.sub(r'\s{2,}', ' ', text).strip()
+
+    # 8) Reattach punctuation to preceding word and normalize post-punct spacing
+    #    "word ." → "word."
+    text = re.sub(r'\s+([\.!,\?;:])', r'\1', text)
+    #    "word.Next" → "word. Next"
+    text = re.sub(r'([\.!,\?;:])(?=\S)', r'\1 ', text)
+
+    return text
+
 def extract_textual_embeddings(text):
     X = hashing_vectorizer.fit_transform([text])
     dense_matrix = X.toarray()
@@ -100,7 +130,7 @@ def compute_and_remove_repeated_ngrams(text, ngram_size=3, threshold=3):
 def calculate_similarity(embed1, embed2):
     return cosine_similarity([embed1], [embed2])[0][0]
 
-def semantic_compress_text(full_text, compression_rate=0.7, num_topics=5, reference_text: str = None):
+def semantic_compress_text(full_text, compression_rate=0.7, num_topics=5, reference_text: str = None, perform_cleaning: bool = True):
     def create_lda_model(texts, stopwords):
         vectorizer = CountVectorizer(stop_words=stopwords)
         doc_term_matrix = vectorizer.fit_transform(texts)
@@ -129,6 +159,9 @@ def semantic_compress_text(full_text, compression_rate=0.7, num_topics=5, refere
         return importance
 
     try:
+        if perform_cleaning:
+            full_text = clean_text(full_text)
+        
         # Split the text into sentences
         sentences = sent_tokenize(full_text)
 
@@ -192,7 +225,7 @@ def semantic_compress_text(full_text, compression_rate=0.7, num_topics=5, refere
     
     return full_text
 
-def compress_text(text, *, target_token_count=None, compression_rate=0.7, reference_text_steering=None):
+def compress_text(text, *, target_token_count=None, compression_rate=0.7, reference_text_steering=None, perform_cleaning=True):
     """
     Compress text using either a compression rate or a target token count.
     If both are provided, the compression rate will be used.
@@ -219,7 +252,8 @@ def compress_text(text, *, target_token_count=None, compression_rate=0.7, refere
         return semantic_compress_text(
             full_text = text,
             compression_rate = compression_rate,
-            reference_text = reference_text_steering
+            reference_text = reference_text_steering,
+            perform_cleaning = perform_cleaning
         )
     except Exception:
         traceback.print_exc()

{semantic_compressor-2.1.dist-info → semantic_compressor-2.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: semantic_compressor
-Version: 2.1
+Version: 2.3
 Author: Carlo Moro
 Author-email: Carlo Moro <cnmoro@gmail.com>
 Classifier: Programming Language :: Python :: 3

{semantic_compressor-2.1.dist-info → semantic_compressor-2.3.dist-info}/RECORD

@@ -1,5 +1,5 @@
 compressor/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-compressor/semantic.py,sha256=0TpOSQBhpfBcqyCs-08NbMOBvdMsvMPvKDlZIz-5Q4Q,15669
+compressor/semantic.py,sha256=36PflgF3qMwEVRWSOgU0IlldvxRZAs9f38EAZuqOk_Y,17065
 compressor/minbpe/__init__.py,sha256=wZ1z2QKkncvGgiZDBc91AP5m7-M-MVenPStKbS6xylE,95
 compressor/minbpe/base.py,sha256=tTKag04RRFnc4ppoieBbDV0V6thzi_ZvZTlhOYIoY7Q,6881
 compressor/minbpe/basic.py,sha256=0kD4tU8l2MZegfPaHMfDo5CnaSzb9i1v9tDBy6GwMbg,2883
@@ -56,6 +56,7 @@ compressor/resources/nltk_data/stemmers/rslp/step5.pt,sha256=1yEVXyDPJvrQE3aQhp1
 compressor/resources/nltk_data/stemmers/rslp/step6.pt,sha256=Gy6SxIzZBd4b2B5yWiNOLUuciNYme9MOcnEM62RJLdw,143
 compressor/resources/nltk_data/taggers/averaged_perceptron_tagger/averaged_perceptron_tagger.pickle,sha256=JaWhnHzteyusODHaW8CvzCw05d0BzU82G7eZlJppYjg,6138625
 compressor/resources/nltk_data/tokenizers/punkt.zip,sha256=UcMHiZSur2UL_I4Ci-T7QrSg0XfUHAErapg5eWU2YOw,13905355
+compressor/resources/nltk_data/tokenizers/punkt_tab.zip,sha256=5X9kGHl0J3cmo0F8pvGB7FQDZ2xxdnLu9qdIp7IOAQY,4319076
 compressor/resources/nltk_data/tokenizers/punkt/.DS_Store,sha256=JDRKA4cwk1i-cV4bgD7JYsmggA_XtgokkqaB24H5GHM,6148
 compressor/resources/nltk_data/tokenizers/punkt/README,sha256=0yUcrmapNZvWjAOeOkYXKgXKnfCyfc36I65ZTWjSfsQ,8574
 compressor/resources/nltk_data/tokenizers/punkt/czech.pickle,sha256=W6c9KTx9eVOVa88C82lexcHw1Sfyo8OAl_VZM5T6FpA,1265552
@@ -97,8 +98,85 @@ compressor/resources/nltk_data/tokenizers/punkt/PY3/slovene.pickle,sha256=Uu8swO
 compressor/resources/nltk_data/tokenizers/punkt/PY3/spanish.pickle,sha256=FkpQ-txaSfjsdCbq4R0xEe51K0ij7zc9R3RQERkqWYQ,562337
 compressor/resources/nltk_data/tokenizers/punkt/PY3/swedish.pickle,sha256=sPfVOL_VJmYzsJ6ELNkungrBDx2SO_IR4Ul5ct3Ecxg,979681
 compressor/resources/nltk_data/tokenizers/punkt/PY3/turkish.pickle,sha256=rmjvWGNyisUzLofrH2uudy_zKhOkyqKwGlxoED6FPFs,1017038
-semantic_compressor-2.1.dist-info/LICENSE,sha256=DFRihXonZ3qVRaTrzuXNaDI_-h2jyT2SqWqjtTDHfqI,1067
-semantic_compressor-2.1.dist-info/METADATA,sha256=-ubUP-0T2LvMpBk3hUG1c429Y-9HuknQX_CciuIbS1c,6178
-semantic_compressor-2.1.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-semantic_compressor-2.1.dist-info/top_level.txt,sha256=qb2SlKrEmMrQDVrhwxu3Wr7U6JupPXtDGrJpIQr8xSc,11
-semantic_compressor-2.1.dist-info/RECORD,,
+compressor/resources/nltk_data/tokenizers/punkt_tab/README,sha256=0yUcrmapNZvWjAOeOkYXKgXKnfCyfc36I65ZTWjSfsQ,8574
+compressor/resources/nltk_data/tokenizers/punkt_tab/czech/abbrev_types.txt,sha256=jrH-taakb5vY9qWuWOIYQopltFm169YWaNg_qArPwyk,462
+compressor/resources/nltk_data/tokenizers/punkt_tab/czech/collocations.tab,sha256=_Vj1SWWMg4m-Mld-IyGHvpN2x5cI9mxgOIqEkWyPPz0,1390
+compressor/resources/nltk_data/tokenizers/punkt_tab/czech/ortho_context.tab,sha256=RRyxpooDuojO0SR-a7SMRC6H3OHJ9GMQRhj6xQmGbNQ,684150
+compressor/resources/nltk_data/tokenizers/punkt_tab/czech/sent_starters.txt,sha256=QvHeGKsp-YQhE7h-wCAgLAwlMWyX03-WhDWnx-4wGMo,352
+compressor/resources/nltk_data/tokenizers/punkt_tab/danish/abbrev_types.txt,sha256=SjWLJre6NyyrPO3rQ7TpmnaunQEVBJOWQUHrQqYuSm4,1130
+compressor/resources/nltk_data/tokenizers/punkt_tab/danish/collocations.tab,sha256=sAPTuDmv1Hr6gdZGIKHPFbdblEISp_PuanVx8EGuAuw,1693
+compressor/resources/nltk_data/tokenizers/punkt_tab/danish/ortho_context.tab,sha256=WdfM2nse72y83bbS8ZMMJVpdQ2AOZuMu_yjquBkMByc,743835
+compressor/resources/nltk_data/tokenizers/punkt_tab/danish/sent_starters.txt,sha256=-N1BvbG7YzV0si04IF4cBzY5oLgMuVkZZL-S2EMXUn8,415
+compressor/resources/nltk_data/tokenizers/punkt_tab/dutch/abbrev_types.txt,sha256=3NSvBxAWksmGBJugerzWt8GiRk705GqTLPktbiIDiN0,434
+compressor/resources/nltk_data/tokenizers/punkt_tab/dutch/collocations.tab,sha256=mgE1Y7zkY3WVptudLusKs_aGhdcEriDZetcYqpGvvko,623
+compressor/resources/nltk_data/tokenizers/punkt_tab/dutch/ortho_context.tab,sha256=4gd7V0hWSpZm5DfAqW6pxb7Q_jq1x1ZTVYR4XTryD8k,428403
+compressor/resources/nltk_data/tokenizers/punkt_tab/dutch/sent_starters.txt,sha256=c0RUrMWVLZGh36yozf2rtxXYzdNXicdJUoY52AdR9Fw,310
+compressor/resources/nltk_data/tokenizers/punkt_tab/english/abbrev_types.txt,sha256=kqPgcPQ9m0xVNHWMpArXNDsE5-Kb_gwutlijlEWk93k,619
+compressor/resources/nltk_data/tokenizers/punkt_tab/english/collocations.tab,sha256=ji2hIl5N0syduiYe4jHMsTSFniG0YAbn9HLF7iaa8M8,594
+compressor/resources/nltk_data/tokenizers/punkt_tab/english/ortho_context.tab,sha256=S7zKJe09PwbAJAKr-EGbnwM7itwG57SC7KTkX4Gl3Ew,236303
+compressor/resources/nltk_data/tokenizers/punkt_tab/english/sent_starters.txt,sha256=8_hTVIPh26SHJBt2SUUWgSO8oyCalkXlms0SJdx27aw,241
+compressor/resources/nltk_data/tokenizers/punkt_tab/estonian/abbrev_types.txt,sha256=AweHj7Mq8eIdjzugZE45yJYC8viqTkMElD9-ng_QgbI,224
+compressor/resources/nltk_data/tokenizers/punkt_tab/estonian/collocations.tab,sha256=8UAKfr88gqPVqsiG61vwahqECUbPBSuxUy2nwlCUeX8,1825
+compressor/resources/nltk_data/tokenizers/punkt_tab/estonian/ortho_context.tab,sha256=dot-biupP_CW4kSvF1w0OxpzRjxKG5m-6-apqlfqHXg,936992
+compressor/resources/nltk_data/tokenizers/punkt_tab/estonian/sent_starters.txt,sha256=et57uMp8UnqsCVF6bvvfT3Semp_A5D0oVsWL6EcDDIM,679
+compressor/resources/nltk_data/tokenizers/punkt_tab/finnish/abbrev_types.txt,sha256=eCm6q8tdaeJiMoOaH1trxtErDlm38KZfarHBY6y-uAA,484
+compressor/resources/nltk_data/tokenizers/punkt_tab/finnish/collocations.tab,sha256=TXF6G00ftxCMqcntLMY_i0gTz69zgZ6W_OeUld3EgpI,3046
+compressor/resources/nltk_data/tokenizers/punkt_tab/finnish/ortho_context.tab,sha256=Q-Vai_qXdWEzegkQXFcANt8RNGxvkeV9hjFSDu-yYWc,1189717
+compressor/resources/nltk_data/tokenizers/punkt_tab/finnish/sent_starters.txt,sha256=q4HPO3dO8GismK25n8s7GgXK8EI736MCvcKF4VWXVn8,810
+compressor/resources/nltk_data/tokenizers/punkt_tab/french/abbrev_types.txt,sha256=AXR-z-0Igw1YLYyUY0tjo22jFsCmf-7ChxzwPw3Vxlc,245
+compressor/resources/nltk_data/tokenizers/punkt_tab/french/collocations.tab,sha256=jd8VDvWg0iaU58TYB_gLbVXacxyoHqE5Yz9WCzeK-ZA,334
+compressor/resources/nltk_data/tokenizers/punkt_tab/french/ortho_context.tab,sha256=cJDvQpJkfU_hNxWb12ilMEAV0GFNWf3PwwsjiWNDZaw,334275
+compressor/resources/nltk_data/tokenizers/punkt_tab/french/sent_starters.txt,sha256=tFoeva6c30bPLovgFrRxi_T4JURfx4PURfrwz7LVafg,233
+compressor/resources/nltk_data/tokenizers/punkt_tab/german/abbrev_types.txt,sha256=JZ5deLydYaEtVbpHVSxOnFygg2TXikZp6GT9behoGOM,246
+compressor/resources/nltk_data/tokenizers/punkt_tab/german/collocations.tab,sha256=93vVE_QugXn1yTIRHDEr41PVkTOH0L_pCpXBCREbyiU,527
+compressor/resources/nltk_data/tokenizers/punkt_tab/german/ortho_context.tab,sha256=AjiH54z7kZ8aAPmpaoj0ITFYJ0jPgz_UaUkO5mWWMHw,943500
+compressor/resources/nltk_data/tokenizers/punkt_tab/german/sent_starters.txt,sha256=_ktj7b3TP3i6CnVfA1aAqGSUixRa6C_JkHJECmKnes0,762
+compressor/resources/nltk_data/tokenizers/punkt_tab/greek/abbrev_types.txt,sha256=DVYlO368g_DyQDR5vPK46lH_3qie5g40AY6zLkdQ1oM,570
+compressor/resources/nltk_data/tokenizers/punkt_tab/greek/collocations.tab,sha256=D9cyU01WF3yAh0ElyolK5gruewUu4oZaQFtFTP1zem8,118
+compressor/resources/nltk_data/tokenizers/punkt_tab/greek/ortho_context.tab,sha256=qH5EZ3kfl0-AjA05b2yCzToBTU-0RYoih-L8vOcITpw,632838
+compressor/resources/nltk_data/tokenizers/punkt_tab/greek/sent_starters.txt,sha256=GkUK7IUWTj-_64FHaMnxIKvXE5cmr1u14qV3CWc5yo4,567
+compressor/resources/nltk_data/tokenizers/punkt_tab/italian/abbrev_types.txt,sha256=XUBva_s15V9r6igr8Ui9zJ5Gd4YOsBqOZ8NLiUJ0rFA,534
+compressor/resources/nltk_data/tokenizers/punkt_tab/italian/collocations.tab,sha256=gWe1igSbsE3o1pLtM9cpY1A6BhcXzTfKyzO7LQBmnK8,53
+compressor/resources/nltk_data/tokenizers/punkt_tab/italian/ortho_context.tab,sha256=Ri_9SLwcm4bv5m0E2CbNHnpXRb6HG4EhdZ7SkgqWc3A,369822
+compressor/resources/nltk_data/tokenizers/punkt_tab/italian/sent_starters.txt,sha256=TcLN_aNNTyhuWHXWtisUh3GnmacfDUW1BNojy2G1nVE,208
+compressor/resources/nltk_data/tokenizers/punkt_tab/malayalam/abbrev_types.txt,sha256=HqLNAWKcjL6sB5nTk4PhDE1pSHNMJV03487aVQumJGQ,4714
+compressor/resources/nltk_data/tokenizers/punkt_tab/malayalam/collocations.tab,sha256=cbw45CEz8ojGYKM4Fp2IWJRBNSgq1BZqLF0jzIoG7XM,3422
+compressor/resources/nltk_data/tokenizers/punkt_tab/malayalam/ortho_context.tab,sha256=UonkP0bvNnn0677ZXvH61bZN16qJVwQq2ytmHMauRCU,118161
+compressor/resources/nltk_data/tokenizers/punkt_tab/malayalam/sent_starters.txt,sha256=kYyQKqnQtycAQQJYJ0rjNdpRFI_uf8JwlatBFLI6JFw,106
+compressor/resources/nltk_data/tokenizers/punkt_tab/norwegian/abbrev_types.txt,sha256=vs7XQHbtq-ZPRe_7gjAS0TlttB81-h2L4qxrab97xA8,496
+compressor/resources/nltk_data/tokenizers/punkt_tab/norwegian/collocations.tab,sha256=nIFbAeL3UYVStT8JASZE3x-g2MdkK6_53F9h6F4L5CI,869
+compressor/resources/nltk_data/tokenizers/punkt_tab/norwegian/ortho_context.tab,sha256=V-oqoSneauQUzQy-y6iyOZxP3ndaIUV3an6PECPz8YM,734070
+compressor/resources/nltk_data/tokenizers/punkt_tab/norwegian/sent_starters.txt,sha256=K0nvDwwuG06LBq7hFy5NiM5q0BSG_kGCz_b8RDpz3es,378
+compressor/resources/nltk_data/tokenizers/punkt_tab/polish/abbrev_types.txt,sha256=Qm6jfKSG0NzDZTZTxhPYU9C7OSkDf47mv2wS3a6HOqE,973
+compressor/resources/nltk_data/tokenizers/punkt_tab/polish/collocations.tab,sha256=lhZT8GBthey4p2DK1Pad8SPwELxLNd4CU_6y14aWZrQ,1101
+compressor/resources/nltk_data/tokenizers/punkt_tab/polish/ortho_context.tab,sha256=93CAQGcBLuVwf_xl2QUjGA4JurCSMKSSUXd76CGhpPY,1066577
+compressor/resources/nltk_data/tokenizers/punkt_tab/polish/sent_starters.txt,sha256=hWQHauIr49VJo_CXoYRYrCln1lvSja18wiPFP1zifRs,518
+compressor/resources/nltk_data/tokenizers/punkt_tab/portuguese/abbrev_types.txt,sha256=zVw8-cTIt2a8ntvJZESw4hibg3smxYOqLM6DoY5ChfU,301
+compressor/resources/nltk_data/tokenizers/punkt_tab/portuguese/collocations.tab,sha256=oVDSyjuL2dh80l8uHSFCYHzfk0Tv6RS9A0r2AXoW2EE,64
+compressor/resources/nltk_data/tokenizers/punkt_tab/portuguese/ortho_context.tab,sha256=LZwUP3syL8dHOKIeCysE_u0WaCRayT5ajlOZ5uDY7Yo,363124
+compressor/resources/nltk_data/tokenizers/punkt_tab/portuguese/sent_starters.txt,sha256=e39KFYax34WSvylMjdDRmq-QdA14zjtAox7_K5Zp0V4,209
+compressor/resources/nltk_data/tokenizers/punkt_tab/russian/abbrev_types.txt,sha256=dRkVJr5RapkYlI_jrUb2wsTe4UMUNZqJm4wuv0lstjE,15363
+compressor/resources/nltk_data/tokenizers/punkt_tab/russian/collocations.tab,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+compressor/resources/nltk_data/tokenizers/punkt_tab/russian/ortho_context.tab,sha256=spbLqRAGVp0-956yz_I3i_W-W3Xp6D6hGbiD08PuabY,10
+compressor/resources/nltk_data/tokenizers/punkt_tab/russian/sent_starters.txt,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+compressor/resources/nltk_data/tokenizers/punkt_tab/slovene/abbrev_types.txt,sha256=sdaxbR0rPAxBMsFSi5hD0VV_f1RARNsn9OjS1phfrys,280
+compressor/resources/nltk_data/tokenizers/punkt_tab/slovene/collocations.tab,sha256=TBDar9_0luNp86iAm-vxh6vIUd9jBFeuk8fKT9o1BlE,1131
+compressor/resources/nltk_data/tokenizers/punkt_tab/slovene/ortho_context.tab,sha256=O6gmvdflVDr3r44xfUZSuotH_YEhV4VaYnTsE4ekaz4,445129
+compressor/resources/nltk_data/tokenizers/punkt_tab/slovene/sent_starters.txt,sha256=ha4U2t7zxMn3U0kdhJGauxDgkAOt_iSHlNSll_q8V5M,358
+compressor/resources/nltk_data/tokenizers/punkt_tab/spanish/abbrev_types.txt,sha256=JGpLPvutRbY-uWcrcYV0tTWs721lbyTcF6RtE6OO-6E,234
+compressor/resources/nltk_data/tokenizers/punkt_tab/spanish/collocations.tab,sha256=-l6n2_YfKxwSGA1mCjASyfqvSWLrElSdd6-V4qQL5l0,74
+compressor/resources/nltk_data/tokenizers/punkt_tab/spanish/ortho_context.tab,sha256=E0bs_FeyYXSYf_qb1KJVimb0ZZKlE9o4WWSHkDUd83A,337707
+compressor/resources/nltk_data/tokenizers/punkt_tab/spanish/sent_starters.txt,sha256=TU9zgLoWOAthWOno6taXPFLyfjyciK9j8VpBxJCZgxk,289
+compressor/resources/nltk_data/tokenizers/punkt_tab/swedish/abbrev_types.txt,sha256=Hhp4vV5_1ufrSxW3UaI9RBzahF1fwLYxlYF3svUE2gA,208
+compressor/resources/nltk_data/tokenizers/punkt_tab/swedish/collocations.tab,sha256=vSJgQzyUYXunz2EC5MEDUJmnLlova2BzRjklLkFjPAQ,136
+compressor/resources/nltk_data/tokenizers/punkt_tab/swedish/ortho_context.tab,sha256=1IxXBdBXCZNijlOcohHwXNtarXrfzG7XskMQlBv6yfg,616904
+compressor/resources/nltk_data/tokenizers/punkt_tab/swedish/sent_starters.txt,sha256=bw96Z3P3SIojuvxsKefUVU55LYnGt6ifvWqDvshe92c,295
+compressor/resources/nltk_data/tokenizers/punkt_tab/turkish/abbrev_types.txt,sha256=3lYHjF7rP-B-JOq2fEqvROIYHL6A7dl2yxYigZWZ1v0,424
+compressor/resources/nltk_data/tokenizers/punkt_tab/turkish/collocations.tab,sha256=BhzimBd2qPh12k8kvr1-E4-NodkFe0PQf1gBSOwQajM,273
+compressor/resources/nltk_data/tokenizers/punkt_tab/turkish/ortho_context.tab,sha256=_CFCJ_mdXqPucNII3xaxmE6rN10ZRu03kGHGz1wXGL4,642682
+compressor/resources/nltk_data/tokenizers/punkt_tab/turkish/sent_starters.txt,sha256=kyOftVtdKubZRahKlOEYuoqBYyaxfNwRuoERvqDJeCg,613
+semantic_compressor-2.3.dist-info/LICENSE,sha256=DFRihXonZ3qVRaTrzuXNaDI_-h2jyT2SqWqjtTDHfqI,1067
+semantic_compressor-2.3.dist-info/METADATA,sha256=eM7GwG1XgI-vBcj9CU4iTUD9fGby4DmsMRQicIuWW0M,6178
+semantic_compressor-2.3.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+semantic_compressor-2.3.dist-info/top_level.txt,sha256=qb2SlKrEmMrQDVrhwxu3Wr7U6JupPXtDGrJpIQr8xSc,11
+semantic_compressor-2.3.dist-info/RECORD,,