Grammalecte  Check-in [0d1b4ca419]

Overview
Comment:[build][core] analyse tokens outside scope
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | core | build | rg
Files: files | file ages | folders
SHA3-256: 0d1b4ca419a31b1d7d8925058a887d4672e363db31b89bdd4f0f2c1487e60677
User & Date: olr on 2018-06-20 18:32:33
Other Links: branch diff | manifest | tags
Context
2018-06-20
19:17
[fr] avoid regex tokens at the beginning of rules check-in: 93b1465f7d user: olr tags: fr, rg
18:32
[build][core] analyse tokens outside scope check-in: 0d1b4ca419 user: olr tags: build, core, rg
18:30
[fr] conversion: regex rules -> graph rules check-in: 52e8c2eb3a user: olr tags: fr, rg
Changes

Modified compile_rules_graph.py from [5fc68e4b07] to [c0410177dd].

11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
...
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
dFUNCTIONS = {}


def prepareFunction (s, bTokenValue=False):
    s = s.replace("__also__", "bCondMemo")
    s = s.replace("__else__", "not bCondMemo")
    s = re.sub(r"(select|exclude|define)[(][\\](\d+)", 'g_\\1(lToken[\\2+nTokenOffset]', s)
    s = re.sub(r"(morph|displayInfo)[(]\\(\d+)", 'g_\\1(lToken[\\2+nTokenOffset]', s)
    s = re.sub(r"(switchGender|has(?:Mas|Fem)Form)[(]\\(\d+)", 'g_\\1(lToken[\\2+nTokenOffset]["sValue"]', s)
    s = re.sub(r"token\(\s*(\d)", 'nextToken(\\1', s)                                       # token(n)
    s = re.sub(r"token\(\s*-(\d)", 'prevToken(\\1', s)                                      # token(-n)
    s = re.sub(r"before\(\s*", 'look(s[:m.start()], ', s)                                   # before(s)
    s = re.sub(r"after\(\s*", 'look(s[m.end():], ', s)                                      # after(s)
    s = re.sub(r"textarea\(\s*", 'look(s, ', s)                                             # textarea(s)
    s = re.sub(r"before_chk1\(\s*", 'look_chk1(dDA, s[:m.start()], 0, ', s)                 # before_chk1(s)
    s = re.sub(r"after_chk1\(\s*", 'look_chk1(dDA, s[m.end():], m.end(), ', s)              # after_chk1(s)
    s = re.sub(r"textarea_chk1\(\s*", 'look_chk1(dDA, s, 0, ', s)                           # textarea_chk1(s)
    s = re.sub(r"\bspell *[(]", '_oSpellChecker.isValid(', s)
................................................................................

    # 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
            sParams = "lToken, nTokenOffset, sCountry, bCondMemo"
        elif sFuncName.startswith("g_m_"): # message
            sParams = "lToken, nTokenOffset"
        elif sFuncName.startswith("g_s_"): # suggestion
            sParams = "lToken, nTokenOffset"
        elif sFuncName.startswith("g_p_"): # preprocessor
            sParams = "lToken"
        elif sFuncName.startswith("g_d_"): # disambiguator







|
|
|
|







 







|







11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
...
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
dFUNCTIONS = {}


def prepareFunction (s, bTokenValue=False):
    s = s.replace("__also__", "bCondMemo")
    s = s.replace("__else__", "not bCondMemo")
    s = re.sub(r"(select|exclude|define)[(][\\](\d+)", 'g_\\1(lToken[\\2+nTokenOffset]', s)
    s = re.sub(r"(morph|analyse|displayInfo)[(]\\(\d+)", 'g_\\1(lToken[\\2+nTokenOffset]', s)
    s = re.sub(r"(switchGender|has(?:Mas|Fem)Form)[(]\\(\d+)", '\\1(lToken[\\2+nTokenOffset]["sValue"]', s)
    s = re.sub(r"(morph|analyse)\(>1", 'g_\\1(lToken[nLastToken+1]', s)                     # next token
    s = re.sub(r"(morph|analyse)\(<1", 'g_\\1(lToken[nTokenOffset]', s)                     # previous token
    s = re.sub(r"before\(\s*", 'look(s[:m.start()], ', s)                                   # before(s)
    s = re.sub(r"after\(\s*", 'look(s[m.end():], ', s)                                      # after(s)
    s = re.sub(r"textarea\(\s*", 'look(s, ', s)                                             # textarea(s)
    s = re.sub(r"before_chk1\(\s*", 'look_chk1(dDA, s[:m.start()], 0, ', s)                 # before_chk1(s)
    s = re.sub(r"after_chk1\(\s*", 'look_chk1(dDA, s[m.end():], m.end(), ', s)              # after_chk1(s)
    s = re.sub(r"textarea_chk1\(\s*", 'look_chk1(dDA, s, 0, ', s)                           # textarea_chk1(s)
    s = re.sub(r"\bspell *[(]", '_oSpellChecker.isValid(', s)
................................................................................

    # 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
            sParams = "lToken, nTokenOffset, nLastToken, sCountry, bCondMemo"
        elif sFuncName.startswith("g_m_"): # message
            sParams = "lToken, nTokenOffset"
        elif sFuncName.startswith("g_s_"): # suggestion
            sParams = "lToken, nTokenOffset"
        elif sFuncName.startswith("g_p_"): # preprocessor
            sParams = "lToken"
        elif sFuncName.startswith("g_d_"): # disambiguator

Modified gc_core/py/lang_core/gc_engine.py from [16c28b42e7] to [5229bc46eb].

704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
                try:
                    if bDebug:
                        print("ACTION:", sRuleId)
                        print(dRule[sRuleId])
                    sOption, sFuncCond, cActionType, sWhat, *eAct = dRule[sRuleId]
                    # action in lActions: [ condition, action type, replacement/suggestion/action[, iTokenStart, iTokenEnd[, nPriority, message, URL]] ]
                    if not sOption or dOptions.get(sOption, False):
                        bCondMemo = not sFuncCond or globals()[sFuncCond](self.lToken, nTokenOffset, sCountry, bCondMemo)
                        if bCondMemo:
                            if cActionType == "-":
                                # grammar error
                                nTokenErrorStart = nTokenOffset + eAct[0]
                                nTokenErrorEnd = (nTokenOffset + eAct[1])  if eAct[1]  else nLastToken
                                nErrorStart = self.nOffsetWithinParagraph + self.lToken[nTokenErrorStart]["nStart"]
                                nErrorEnd = self.nOffsetWithinParagraph + self.lToken[nTokenErrorEnd]["nEnd"]







|







704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
                try:
                    if bDebug:
                        print("ACTION:", sRuleId)
                        print(dRule[sRuleId])
                    sOption, sFuncCond, cActionType, sWhat, *eAct = dRule[sRuleId]
                    # action in lActions: [ condition, action type, replacement/suggestion/action[, iTokenStart, iTokenEnd[, nPriority, message, URL]] ]
                    if not sOption or dOptions.get(sOption, False):
                        bCondMemo = not sFuncCond or globals()[sFuncCond](self.lToken, nTokenOffset, nLastToken, sCountry, bCondMemo)
                        if bCondMemo:
                            if cActionType == "-":
                                # grammar error
                                nTokenErrorStart = nTokenOffset + eAct[0]
                                nTokenErrorEnd = (nTokenOffset + eAct[1])  if eAct[1]  else nLastToken
                                nErrorStart = self.nOffsetWithinParagraph + self.lToken[nTokenErrorStart]["nStart"]
                                nErrorEnd = self.nOffsetWithinParagraph + self.lToken[nTokenErrorEnd]["nEnd"]