factored out some common util code

jtauber · Dec 1, 2016 · e0b3614 · e0b3614
1 parent 6c7a4a5
commit e0b3614
Show file tree

Hide file tree

Showing 4 changed files with 82 additions and 151 deletions.
diff --git a/generate_morphgnt_lexicon.py b/generate_morphgnt_lexicon.py
@@ -6,6 +6,7 @@
 
 from accent import strip_length
 from greek_inflexion import GreekInflexion
+from morphgnt_utils import bcv_tuple, convert_parse, key_to_part
 
 
 ginflexion = GreekInflexion(
@@ -15,84 +16,6 @@
 )
 
 
-# @@@ move this to greek-utils
-
-def bcv_tuple(bcv):
-    """
-    converts a BBCCVV string into a tuple of book, chapter, verse number.
-
-    e.g. "012801" returns (1, 28, 1)
-    """
-    return (int(i) for i in [bcv[0:2], bcv[2:4], bcv[4:6]])
-
-
-def convert_parse(ccat_parse):
-    if ccat_parse[3] in "DISO":
-        result = ccat_parse[1:4] + "." + ccat_parse[0] + ccat_parse[5]
-    elif ccat_parse[3] == "P":
-        result = ccat_parse[1:4] + "." + ccat_parse[4:7]
-    elif ccat_parse[3] == "N":
-        result = ccat_parse[1:4]
-    if result[1] == "P" and result[0] not in "AF":
-        result = result[0] + "M" + result[2:]
-    return result
-
-
-PARTS = {
-    "1-": [
-        "PAD", "PAI", "PAN", "PAP", "PAS", "PAO",
-        "PMD", "PMI", "PMN", "PMP", "PMS",
-    ],
-    "1+": [
-        "IAI",
-        "IMI",
-    ],
-    "2-": [
-        "FAI", "FAN", "FAP",
-        "FMI",
-    ],
-    "3-": [
-        "AAD", "AAN", "AAP", "AAS", "AAO",
-        "AMD", "AMN", "AMP", "AMS", "AMO",
-    ],
-    "3+": [
-        "AAI",
-        "AMI",
-    ],
-    "4-": [
-        "XAI", "XAN", "XAP", "XAS",
-    ],
-    "4+": [
-        "YAI",
-    ],
-    "5-": [
-        "XMD", "XMI", "XMN", "XMP",
-    ],
-    "5+": [
-        "YMI",
-    ],
-    "6-": [
-        "APD", "APN", "APP", "APS", "APO",
-    ],
-    "6+": [
-        "API",
-    ],
-    "7-": [
-        "FPI", "FPP",
-    ]
-}
-
-REVERSE_PARTS = {}
-
-for part, tvm_list in PARTS.items():
-    for tvm in tvm_list:
-        REVERSE_PARTS[tvm] = part
-
-
-def pp(key):
-    return REVERSE_PARTS[key[0:3]]
-
-
 STEM_GUESSES = defaultdict(lambda: defaultdict(set))
 
 for book_num in [4, 23, 24, 25]:
@@ -123,7 +46,8 @@ def pp(key):
                 correct = "✕"
             if correct == "✕":
                 if stem_guess:
-                    STEM_GUESSES[lemma][pp(key)].add(frozenset(stem_guess))
+                    STEM_GUESSES[lemma][key_to_part(key)].add(
+                        frozenset(stem_guess))
 
 
 for lemma, parts in sorted(STEM_GUESSES.items()):

diff --git a/make_morphgnt_test.py b/make_morphgnt_test.py
@@ -4,51 +4,10 @@
 
 from pysblgnt import morphgnt_rows
 
-BOOK_NUM = 4
-
-
-# @@@ move this to greek-utils
-
-def bcv_tuple(bcv):
-    """
-    converts a BBCCVV string into a tuple of book, chapter, verse number.
+from morphgnt_utils import bcv_tuple, convert_parse
 
-    e.g. "012801" returns (1, 28, 1)
-    """
-    return (int(i) for i in [bcv[0:2], bcv[2:4], bcv[4:6]])
 
-
-PARSE_CODES = [
-    ('-AAPGSM-', 'V-AAP-GSM'),
-    ('-PAPNPF-', 'V-PAP-NPF'),
-    ('-PMPNPF-', 'V-PNP-NPF'),
-    ('-XAPNPM-', 'V-RAP-NPM'),
-    ('-XPPASN-', 'V-RPP-ASN'),
-    ('2AAD-P--', 'V-AAM-2P'),
-    ('2PAD-P--', 'V-PAM-2P'),
-    ('2XAI-S--', 'V-RAI-2S'),
-    ('3AAI-P--', 'V-AAI-3P'),
-    ('3AAI-S--', 'V-AAI-3S'),
-    ('3AMI-S--', 'V-2ADI-3S'),  # @@@
-    ('3AMI-S--', 'V-ADI-3S'),
-    ('3API-S--', 'V-API-3S'),
-    ('3APS-P--', 'V-APS-3P'),
-    ('3IAI-P--', 'V-IAI-3P'),
-    ('3IAI-S--', 'V-IAI-3S'),
-    ('3PAI-P--', 'V-PAI-3P'),
-    ('3PAI-S--', 'V-PAI-3S'),
-    ('3PAS-S--', 'V-PAS-3S'),
-    ('3YAI-P--', 'V-LAI-3P'),
-    ('3YAI-S--', 'V-LAI-3S'),
-]
-
-
-def convert_parse(ccat_parse):
-    if ccat_parse[3] in "DISO":
-        result = ccat_parse[1:4] + "." + ccat_parse[0] + ccat_parse[5]
-    elif ccat_parse[3] == "P":
-        result = ccat_parse[1:4] + "." + ccat_parse[4:7]
-    return result
+BOOK_NUM = 4
 
 
 VERBS = defaultdict(lambda: defaultdict(set))
@@ -57,11 +16,6 @@ def convert_parse(ccat_parse):
     b, c, v = bcv_tuple(row["bcv"])
     if c == 2 and v <= 11:
         if row["ccat-pos"] == "V-":
-            assert (
-                row["ccat-parse"], row["robinson"]
-            ) in PARSE_CODES, (
-                row["ccat-parse"], row["robinson"]
-            )
             VERBS[row["lemma"]][convert_parse(row["ccat-parse"])].add(
                 row["norm"])
 

diff --git a/morphgnt_generate.py b/morphgnt_generate.py
@@ -4,36 +4,13 @@
 
 from accent import strip_length  # , rebreath
 from greek_inflexion import GreekInflexion
-
+from morphgnt_utils import bcv_tuple, convert_parse
 
 ginflexion = GreekInflexion("stemming.yaml", "morphgnt_johannine_lexicon.yaml")
 
 debug = False
 
 
-# @@@ move this to greek-utils
-
-def bcv_tuple(bcv):
-    """
-    converts a BBCCVV string into a tuple of book, chapter, verse number.
-
-    e.g. "012801" returns (1, 28, 1)
-    """
-    return (int(i) for i in [bcv[0:2], bcv[2:4], bcv[4:6]])
-
-
-def convert_parse(ccat_parse):
-    if ccat_parse[3] in "DISO":
-        result = ccat_parse[1:4] + "." + ccat_parse[0] + ccat_parse[5]
-    elif ccat_parse[3] == "P":
-        result = ccat_parse[1:4] + "." + ccat_parse[4:7]
-    elif ccat_parse[3] == "N":
-        result = ccat_parse[1:4]
-    if result[1] == "P" and result[0] not in "AF":
-        result = result[0] + "M" + result[2:]
-    return result
-
-
 incorrect_count = 0
 total_count = 0
 

diff --git a/morphgnt_utils.py b/morphgnt_utils.py
@@ -0,0 +1,76 @@
+# @@@ move this to greek-utils
+
+def bcv_tuple(bcv):
+    """
+    converts a BBCCVV string into a tuple of book, chapter, verse number.
+
+    e.g. "012801" returns (1, 28, 1)
+    """
+    return (int(i) for i in [bcv[0:2], bcv[2:4], bcv[4:6]])
+
+
+def convert_parse(ccat_parse):
+    if ccat_parse[3] in "DISO":
+        result = ccat_parse[1:4] + "." + ccat_parse[0] + ccat_parse[5]
+    elif ccat_parse[3] == "P":
+        result = ccat_parse[1:4] + "." + ccat_parse[4:7]
+    elif ccat_parse[3] == "N":
+        result = ccat_parse[1:4]
+    if result[1] == "P" and result[0] not in "AF":
+        result = result[0] + "M" + result[2:]
+    return result
+
+
+PARTS = {
+    "1-": [
+        "PAD", "PAI", "PAN", "PAP", "PAS", "PAO",
+        "PMD", "PMI", "PMN", "PMP", "PMS",
+    ],
+    "1+": [
+        "IAI",
+        "IMI",
+    ],
+    "2-": [
+        "FAI", "FAN", "FAP",
+        "FMI",
+    ],
+    "3-": [
+        "AAD", "AAN", "AAP", "AAS", "AAO",
+        "AMD", "AMN", "AMP", "AMS", "AMO",
+    ],
+    "3+": [
+        "AAI",
+        "AMI",
+    ],
+    "4-": [
+        "XAI", "XAN", "XAP", "XAS",
+    ],
+    "4+": [
+        "YAI",
+    ],
+    "5-": [
+        "XMD", "XMI", "XMN", "XMP",
+    ],
+    "5+": [
+        "YMI",
+    ],
+    "6-": [
+        "APD", "APN", "APP", "APS", "APO",
+    ],
+    "6+": [
+        "API",
+    ],
+    "7-": [
+        "FPI", "FPP",
+    ]
+}
+
+REVERSE_PARTS = {}
+
+for part, tvm_list in PARTS.items():
+    for tvm in tvm_list:
+        REVERSE_PARTS[tvm] = part
+
+
+def key_to_part(key):
+    return REVERSE_PARTS[key[0:3]]