Grammalecte  Check-in [9d39408049]

Overview
Comment:[build] fix rules generation when first token was multiple and a selected group
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | build | rg
Files: files | file ages | folders
SHA3-256: 9d39408049f4ddf774c119390c6817e591ac55d6bfd86e023dd79f1442e587c7
User & Date: olr on 2018-07-15 11:33:00
Other Links: branch diff | manifest | tags
Context
2018-07-16
13:37
[build] changeReferenceToken before prepareFunction check-in: 7d80af29a6 user: olr tags: build, rg
2018-07-15
11:33
[build] fix rules generation when first token was multiple and a selected group check-in: 9d39408049 user: olr tags: build, rg
09:43
[build] prepareFunctions: use token value by default check-in: 450e1ec422 user: olr tags: build, rg
Changes

Modified compile_rules_graph.py from [58681b66b6] to [60e97fc2e5].

19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
..
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
...
100
101
102
103
104
105
106


107
108
109
110
111
112
113
...
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
...
345
346
347
348
349
350
351

352
353
354
355
356


357
358
359
360
361
362
363
364
365
    s = s.replace("__also__", "bCondMemo")
    s = s.replace("__else__", "not bCondMemo")
    s = re.sub(r"(morph|analyse|value|displayInfo)[(]\\(\d+)", 'g_\\1(lToken[\\2+nTokenOffset]', s)
    s = re.sub(r"(select|exclude|define|define_from)[(][\\](\d+)", 'g_\\1(lToken[\\2+nTokenOffset]', s)
    s = re.sub(r"(tag_before|tag_after)[(][\\](\d+)", 'g_\\1(lToken[\\2+nTokenOffset], dTags', s)
    s = re.sub(r"space_after[(][\\](\d+)", 'g_space_between_tokens(lToken[\\1+nTokenOffset], lToken[\\1+nTokenOffset+1]', s)
    s = re.sub(r"analyse_with_next[(][\\](\d+)", 'g_merged_analyse(lToken[\\1+nTokenOffset], lToken[\\1+nTokenOffset+1]', s)
    #s = re.sub(r"(switchGender|has(?:(?:Mas|Fem)Form)|Simil)[(]\\(\d+)", '\\1(lToken[\\2+nTokenOffset]["sValue"]', s)
    s = re.sub(r"(morph|analyse|value)\(>1", 'g_\\1(lToken[nLastToken+1]', s)                       # next token
    s = re.sub(r"(morph|analyse|value)\(<1", 'g_\\1(lToken[nTokenOffset]', s)                       # previous token
    #s = re.sub(r"[\\](\d+)\.is(upper|lower|title)\(\)", 'lToken[\\1+nTokenOffset]["sValue"].is\\2()', s)
    #s = re.sub(r"[\\](\d+)\.(startswith|endswith)\(", 'lToken[\\1+nTokenOffset]["sValue"].\\2(', s)
    s = re.sub(r"\bspell *[(]", '_oSpellChecker.isValid(', s)
    s = re.sub(r"\bbefore\(\s*", 'look(sSentence[:lToken[1+nTokenOffset]["nStart"]], ', s)          # before(s)
    s = re.sub(r"\bafter\(\s*", 'look(sSentence[lToken[nLastToken]["nEnd"]:], ', s)                 # after(s)
    s = re.sub(r"\bbefore0\(\s*", 'look(sSentence0[:lToken[1+nTokenOffset]["nStart"]], ', s)        # before0(s)
    s = re.sub(r"\bafter0\(\s*", 'look(sSentence[lToken[nLastToken]["nEnd"]:], ', s)                # after0(s)
    s = re.sub(r"[\\](\d+)", 'lToken[\\1+nTokenOffset]["sValue"]', s)
    return s
................................................................................
        if ( (sToken.startswith("[") and sToken.endswith("]")) or (sToken.startswith("([") and sToken.endswith("])")) ):
            # multiple token
            bSelectedGroup = sToken.startswith("(") and sToken.endswith(")")
            if bSelectedGroup:
                sToken = sToken[1:-1]
            lNewToken = sToken[1:-1].split("|")
            if not lTokenLines:
                lTokenLines = [ [s]  for s  in lNewToken ]
                if bNullPossible:
                    lTokenLines.extend([ []  for i  in range(len(lNewToken)+1) ])
            else:
                lNewTemp = []
                if bNullPossible:
                    for aRule in lTokenLines:
                        for sElem in lNewToken:
................................................................................
def createRule (iLine, sRuleName, sTokenLine, iActionBlock, sActions, nPriority, dOptPriority, dDef):
    "generator: create rule as list"
    # print(iLine, "//", sRuleName, "//", sTokenLine, "//", sActions, "//", nPriority)
    for lToken in genTokenLines(sTokenLine, dDef):
        # Calculate positions
        dPos = {}   # key: iGroup, value: iToken
        iGroup = 0


        for i, sToken in enumerate(lToken):
            if sToken.startswith("(") and sToken.endswith(")"):
                lToken[i] = sToken[1:-1]
                iGroup += 1
                dPos[iGroup] = i + 1    # we add 1, for we count tokens from 1 to n (not from 0)

        # Parse actions
................................................................................
                sRuleName = m.group(1)
                iActionBlock = 1
                nPriority = int(m.group(2)[1:]) if m.group(2)  else -1
            else:
                print("Error at rule group: ", sLine, " -- line:", i)
                break
        elif re.search("^    +<<- ", sLine) or sLine.startswith("        ") \
                or re.search("^    +#", sLine) or re.search(r"^    [-~=>/](?:\d(?::\d+|)|)>> ", sLine) :
            # actions
            sActions += " " + sLine.strip()
        elif re.match("[  ]*$", sLine):
            # empty line to end merging
            if not lTokenLine:
                continue
            if not sActions:
................................................................................
        for i, sRuleGroup, sTokenLine, iActionBlock, sActions, nPriority in lRuleLine:
            for lRule in createRule(i, sRuleGroup, sTokenLine, iActionBlock, sActions, nPriority, dOptPriority, dDef):
                lPreparedRule.append(lRule)
        # Graph creation
        oDARG = darg.DARG(lPreparedRule, sLang)
        dAllGraph[sGraphName] = oDARG.createGraph()
        # Debugging

        #print("\nGRAPH:", sGraphName)
        #for e in lPreparedRule:
        #    if e[-2] == "##4239":
        #        print(e)
        #    print(e)


        #for k, v in dAllGraph[sGraphName].items():
        #    print(k, "\t", v)

    # creating file with all functions callable by rules
    print("  creating callables...")
    sPyCallables = "# generated code, do not edit\n"
    #sJSCallables = "// generated code, do not edit\nconst oEvalFunc = {\n"
    for sFuncName, sReturn in dFUNCTIONS.items():
        if sFuncName.startswith("_g_c_"): # condition







<


<
<







 







|







 







>
>







 







|







 







>
|
|
|
<
|
>
>
|
|







19
20
21
22
23
24
25

26
27


28
29
30
31
32
33
34
..
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
..
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
...
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
...
344
345
346
347
348
349
350
351
352
353
354

355
356
357
358
359
360
361
362
363
364
365
366
    s = s.replace("__also__", "bCondMemo")
    s = s.replace("__else__", "not bCondMemo")
    s = re.sub(r"(morph|analyse|value|displayInfo)[(]\\(\d+)", 'g_\\1(lToken[\\2+nTokenOffset]', s)
    s = re.sub(r"(select|exclude|define|define_from)[(][\\](\d+)", 'g_\\1(lToken[\\2+nTokenOffset]', s)
    s = re.sub(r"(tag_before|tag_after)[(][\\](\d+)", 'g_\\1(lToken[\\2+nTokenOffset], dTags', s)
    s = re.sub(r"space_after[(][\\](\d+)", 'g_space_between_tokens(lToken[\\1+nTokenOffset], lToken[\\1+nTokenOffset+1]', s)
    s = re.sub(r"analyse_with_next[(][\\](\d+)", 'g_merged_analyse(lToken[\\1+nTokenOffset], lToken[\\1+nTokenOffset+1]', s)

    s = re.sub(r"(morph|analyse|value)\(>1", 'g_\\1(lToken[nLastToken+1]', s)                       # next token
    s = re.sub(r"(morph|analyse|value)\(<1", 'g_\\1(lToken[nTokenOffset]', s)                       # previous token


    s = re.sub(r"\bspell *[(]", '_oSpellChecker.isValid(', s)
    s = re.sub(r"\bbefore\(\s*", 'look(sSentence[:lToken[1+nTokenOffset]["nStart"]], ', s)          # before(s)
    s = re.sub(r"\bafter\(\s*", 'look(sSentence[lToken[nLastToken]["nEnd"]:], ', s)                 # after(s)
    s = re.sub(r"\bbefore0\(\s*", 'look(sSentence0[:lToken[1+nTokenOffset]["nStart"]], ', s)        # before0(s)
    s = re.sub(r"\bafter0\(\s*", 'look(sSentence[lToken[nLastToken]["nEnd"]:], ', s)                # after0(s)
    s = re.sub(r"[\\](\d+)", 'lToken[\\1+nTokenOffset]["sValue"]', s)
    return s
................................................................................
        if ( (sToken.startswith("[") and sToken.endswith("]")) or (sToken.startswith("([") and sToken.endswith("])")) ):
            # multiple token
            bSelectedGroup = sToken.startswith("(") and sToken.endswith(")")
            if bSelectedGroup:
                sToken = sToken[1:-1]
            lNewToken = sToken[1:-1].split("|")
            if not lTokenLines:
                lTokenLines = [ ["("+s+")"]  for s  in lNewToken ]  if bSelectedGroup  else [ [s]  for s  in lNewToken ]
                if bNullPossible:
                    lTokenLines.extend([ []  for i  in range(len(lNewToken)+1) ])
            else:
                lNewTemp = []
                if bNullPossible:
                    for aRule in lTokenLines:
                        for sElem in lNewToken:
................................................................................
def createRule (iLine, sRuleName, sTokenLine, iActionBlock, sActions, nPriority, dOptPriority, dDef):
    "generator: create rule as list"
    # print(iLine, "//", sRuleName, "//", sTokenLine, "//", sActions, "//", nPriority)
    for lToken in genTokenLines(sTokenLine, dDef):
        # Calculate positions
        dPos = {}   # key: iGroup, value: iToken
        iGroup = 0
        if iLine == 2211:
            print(lToken)
        for i, sToken in enumerate(lToken):
            if sToken.startswith("(") and sToken.endswith(")"):
                lToken[i] = sToken[1:-1]
                iGroup += 1
                dPos[iGroup] = i + 1    # we add 1, for we count tokens from 1 to n (not from 0)

        # Parse actions
................................................................................
                sRuleName = m.group(1)
                iActionBlock = 1
                nPriority = int(m.group(2)[1:]) if m.group(2)  else -1
            else:
                print("Error at rule group: ", sLine, " -- line:", i)
                break
        elif re.search("^    +<<- ", sLine) or sLine.startswith("        ") \
                or re.search("^    +#", sLine) or re.search(r"^    [-~=>/](?:\d\.?(?::\.?\d+|)|)>> ", sLine) :
            # actions
            sActions += " " + sLine.strip()
        elif re.match("[  ]*$", sLine):
            # empty line to end merging
            if not lTokenLine:
                continue
            if not sActions:
................................................................................
        for i, sRuleGroup, sTokenLine, iActionBlock, sActions, nPriority in lRuleLine:
            for lRule in createRule(i, sRuleGroup, sTokenLine, iActionBlock, sActions, nPriority, dOptPriority, dDef):
                lPreparedRule.append(lRule)
        # Graph creation
        oDARG = darg.DARG(lPreparedRule, sLang)
        dAllGraph[sGraphName] = oDARG.createGraph()
        # Debugging
        if False:
            print("\nRULES:")
            for e in lPreparedRule:
                if e[-2] == "##2211":

                    print(e)
        if False:
            print("\nGRAPH:", sGraphName)
            for k, v in dAllGraph[sGraphName].items():
                print(k, "\t", v)

    # creating file with all functions callable by rules
    print("  creating callables...")
    sPyCallables = "# generated code, do not edit\n"
    #sJSCallables = "// generated code, do not edit\nconst oEvalFunc = {\n"
    for sFuncName, sReturn in dFUNCTIONS.items():
        if sFuncName.startswith("_g_c_"): # condition