Grammalecte  Check-in [5d4721e754]

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:[build] variable renaming and line id instead of rule id to identify rules
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk | build
Files: files | file ages | folders
SHA3-256: 5d4721e754a2fa3a73f2198b2b25c67f8ac7eef820872c8e43ea73f6fa4397ab
User & Date: olr 2020-06-15 07:43:47
Context
2020-06-15
12:49
[fr] ajustements: typographie check-in: fc1bc3e152 user: olr tags: fr, trunk
07:43
[build] variable renaming and line id instead of rule id to identify rules check-in: 5d4721e754 user: olr tags: build, trunk
07:29
[fr] ajustements check-in: efb4fbf4aa user: olr tags: fr, trunk
Changes

Changes to compile_rules.py.

131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
...
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
        return re.compile(sRegex).groups
    except re.error:
        traceback.print_exc()
        print(sRegex)
    return 0


def createRule (s, nIdLine, sLang, bParagraph, dOptPriority):
    "returns rule as list [option name, regex, bCaseInsensitive, identifier, list of actions]"
    global dJSREGEXES

    sLineId = f"#{nIdLine}" + ("p" if bParagraph else "s")
    sRuleId = sLineId

    #### GRAPH CALL
    if s.startswith("@@@@"):
        if bParagraph:
            print("Error. Graph call can be made only after the first pass (sentence by sentence)")
            exit()
................................................................................
    for _ in re.finditer(r"\(\?:[^)]*\([\[\w -]", sRegex):
        print(f"# Warning: groups inside non grouping parenthesis in regex at line {sLineId}")

    #### PARSE ACTIONS
    lActions = []
    nAction = 1
    for sAction in s.split(" <<- "):
        t = createAction(sRuleId + "_" + str(nAction), sAction, nGroup)
        nAction += 1
        if t:
            lActions.append(t)
    if not lActions:
        return None

    return [sOption, sRegex, bCaseInsensitive, sLineId, sRuleId, nPriority, lActions, tGroups]


def checkReferenceNumbers (sText, sActionId, nToken):
    "check if token references in <sText> greater than <nToken> (debugging)"
    for x in re.finditer(r"\\(\d+)", sText):
        if int(x.group(1)) > nToken:
            print(f"# Error in token index at line {sActionId} ({nToken} tokens only)")
            print(sText)


def checkIfThereIsCode (sText, sActionId):
    "check if there is code in <sText> (debugging)"
    if re.search("[.]\\w+[(]|sugg\\w+[(]|\\([0-9]|\\[[0-9]", sText):
        print(f"# Warning at line {sActionId}:  This message looks like code. Line should probably begin with =")
        print(sText)


def createAction (sIdAction, sAction, nGroup):
    "returns an action to perform as a tuple (condition, action type, action[, iGroup [, message, URL ]])"
    m = re.search(r"([-~=>])(\d*|)>>", sAction)
    if not m:
        print(f"# No action at line {sIdAction}")
        return None

    #### CONDITION
    sCondition = sAction[:m.start()].strip()
    if sCondition:
        sCondition = prepareFunction(sCondition)
        lFUNCTIONS.append(("_c_"+sIdAction, sCondition))
        checkReferenceNumbers(sCondition, sIdAction, nGroup)
        if ".match" in sCondition:
            print("# Error. JS compatibility. Don't use .match() in condition, use .search()")
        sCondition = "_c_"+sIdAction
    else:
        sCondition = None

    #### iGroup / positioning
    iGroup = int(m.group(2)) if m.group(2) else 0
    if iGroup > nGroup:
        print(f"# Selected group > group number in regex at line {sIdAction}")

    #### ACTION
    sAction = sAction[m.end():].strip()
    cAction = m.group(1)
    if cAction == "-":
        ## error
        iMsg = sAction.find(" && ")
        if iMsg == -1:
            sMsg = "# Error. Error message not found."
            sURL = ""
            print(f"# No message. Action id: {sIdAction}")
        else:
            sMsg = sAction[iMsg+4:].strip()
            sAction = sAction[:iMsg].strip()
            sURL = ""
            mURL = re.search("[|] *(https?://.*)", sMsg)
            if mURL:
                sURL = mURL.group(1).strip()
                sMsg = sMsg[:mURL.start(0)].strip()
            checkReferenceNumbers(sMsg, sIdAction, nGroup)
            if sMsg[0:1] == "=":
                sMsg = prepareFunction(sMsg[1:])
                lFUNCTIONS.append(("_m_"+sIdAction, sMsg))
                sMsg = "=_m_"+sIdAction
            else:
                checkIfThereIsCode(sMsg, sIdAction)

    checkReferenceNumbers(sAction, sIdAction, nGroup)
    if sAction[0:1] == "=" or cAction == "=":
        sAction = prepareFunction(sAction)
        sAction = sAction.replace("m.group(i[4])", "m.group("+str(iGroup)+")")
    else:
        checkIfThereIsCode(sAction, sIdAction)

    if cAction == ">":
        ## no action, break loop if condition is False
        return [sCondition, cAction, ""]

    if not sAction:
        print(f"# Error in action at line {sIdAction}:  This action is empty.")
        return None

    if cAction == "-":
        ## error detected --> suggestion
        if sAction[0:1] == "=":
            lFUNCTIONS.append(("_s_"+sIdAction, sAction[1:]))
            sAction = "=_s_"+sIdAction
        elif sAction.startswith('"') and sAction.endswith('"'):
            sAction = sAction[1:-1]
        if not sMsg:
            print(f"# Error in action at line {sIdAction}:  the message is empty.")
        return [sCondition, cAction, sAction, iGroup, sMsg, sURL]
    if cAction == "~":
        ## text processor
        if sAction[0:1] == "=":
            lFUNCTIONS.append(("_p_"+sIdAction, sAction[1:]))
            sAction = "=_p_"+sIdAction
        elif sAction.startswith('"') and sAction.endswith('"'):
            sAction = sAction[1:-1]
        return [sCondition, cAction, sAction, iGroup]
    if cAction == "=":
        ## disambiguator
        if sAction[0:1] == "=":
            sAction = sAction[1:]
        lFUNCTIONS.append(("_d_"+sIdAction, sAction))
        sAction = "_d_"+sIdAction
        return [sCondition, cAction, sAction]
    print(f"# Unknown action at line {sIdAction}")
    return None


def _calcRulesStats (lRules):
    "count rules and actions"
    d = {'=':0, '~': 0, '-': 0, '>': 0}
    for aRule in lRules:







|



|







 







|













|



|


|



|



|






|
|

|
|






|










|








|


|
|

|

|




|






|





|
|



|




|
|







|
|

|







131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
...
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
        return re.compile(sRegex).groups
    except re.error:
        traceback.print_exc()
        print(sRegex)
    return 0


def createRule (s, nLineId, sLang, bParagraph, dOptPriority):
    "returns rule as list [option name, regex, bCaseInsensitive, identifier, list of actions]"
    global dJSREGEXES

    sLineId = f"#{nLineId}" + ("p" if bParagraph else "s")
    sRuleId = sLineId

    #### GRAPH CALL
    if s.startswith("@@@@"):
        if bParagraph:
            print("Error. Graph call can be made only after the first pass (sentence by sentence)")
            exit()
................................................................................
    for _ in re.finditer(r"\(\?:[^)]*\([\[\w -]", sRegex):
        print(f"# Warning: groups inside non grouping parenthesis in regex at line {sLineId}")

    #### PARSE ACTIONS
    lActions = []
    nAction = 1
    for sAction in s.split(" <<- "):
        t = createAction(sLineId, sRuleId + "_" + str(nAction), sAction, nGroup)
        nAction += 1
        if t:
            lActions.append(t)
    if not lActions:
        return None

    return [sOption, sRegex, bCaseInsensitive, sLineId, sRuleId, nPriority, lActions, tGroups]


def checkReferenceNumbers (sText, sActionId, nToken):
    "check if token references in <sText> greater than <nToken> (debugging)"
    for x in re.finditer(r"\\(\d+)", sText):
        if int(x.group(1)) > nToken:
            print(f"# Error in token index at line {sLineId} / {sActionId} ({nToken} tokens only)")
            print(sText)


def checkIfThereIsCode (sText, sLineId, sActionId):
    "check if there is code in <sText> (debugging)"
    if re.search("[.]\\w+[(]|sugg\\w+[(]|\\([0-9]|\\[[0-9]", sText):
        print(f"# Warning at line {sLineId} / {sActionId}:  This message looks like code. Line should probably begin with =")
        print(sText)


def createAction (sLineId, sActionId, sAction, nGroup):
    "returns an action to perform as a tuple (condition, action type, action[, iGroup [, message, URL ]])"
    m = re.search(r"([-~=>])(\d*|)>>", sAction)
    if not m:
        print(f"# No action at line {sLineId} / {sActionId}")
        return None

    #### CONDITION
    sCondition = sAction[:m.start()].strip()
    if sCondition:
        sCondition = prepareFunction(sCondition)
        lFUNCTIONS.append(("_c_"+sActionId, sCondition))
        checkReferenceNumbers(sCondition, sActionId, nGroup)
        if ".match" in sCondition:
            print(f"# Error at line {sLineId} / {sActionId}. JS compatibility. Don't use .match() in condition, use .search()")
        sCondition = "_c_"+sActionId
    else:
        sCondition = None

    #### iGroup / positioning
    iGroup = int(m.group(2)) if m.group(2) else 0
    if iGroup > nGroup:
        print(f"# Error. Selected group > group number in regex at line {sLineId} / {sActionId}")

    #### ACTION
    sAction = sAction[m.end():].strip()
    cAction = m.group(1)
    if cAction == "-":
        ## error
        iMsg = sAction.find(" && ")
        if iMsg == -1:
            sMsg = "# Error. Error message not found."
            sURL = ""
            print(f"# No message. Id: {sLineId} / {sActionId}")
        else:
            sMsg = sAction[iMsg+4:].strip()
            sAction = sAction[:iMsg].strip()
            sURL = ""
            mURL = re.search("[|] *(https?://.*)", sMsg)
            if mURL:
                sURL = mURL.group(1).strip()
                sMsg = sMsg[:mURL.start(0)].strip()
            checkReferenceNumbers(sMsg, sActionId, nGroup)
            if sMsg[0:1] == "=":
                sMsg = prepareFunction(sMsg[1:])
                lFUNCTIONS.append(("_m_"+sActionId, sMsg))
                sMsg = "=_m_"+sActionId
            else:
                checkIfThereIsCode(sMsg, sLineId, sActionId)

    checkReferenceNumbers(sAction, sActionId, nGroup)
    if sAction[0:1] == "=" or cAction == "=":
        sAction = prepareFunction(sAction)
        sAction = sAction.replace("m.group(i[4])", "m.group("+str(iGroup)+")")
    else:
        checkIfThereIsCode(sAction, sLineId, sActionId)

    if cAction == ">":
        ## no action, break loop if condition is False
        return [sCondition, cAction, ""]

    if not sAction:
        print(f"# Error in action at line {sLineId} / {sActionId}:  This action is empty.")
        return None

    if cAction == "-":
        ## error detected --> suggestion
        if sAction[0:1] == "=":
            lFUNCTIONS.append(("_s_"+sActionId, sAction[1:]))
            sAction = "=_s_"+sActionId
        elif sAction.startswith('"') and sAction.endswith('"'):
            sAction = sAction[1:-1]
        if not sMsg:
            print(f"# Error in action at line {sLineId} / {sActionId}:  the message is empty.")
        return [sCondition, cAction, sAction, iGroup, sMsg, sURL]
    if cAction == "~":
        ## text processor
        if sAction[0:1] == "=":
            lFUNCTIONS.append(("_p_"+sActionId, sAction[1:]))
            sAction = "=_p_"+sActionId
        elif sAction.startswith('"') and sAction.endswith('"'):
            sAction = sAction[1:-1]
        return [sCondition, cAction, sAction, iGroup]
    if cAction == "=":
        ## disambiguator
        if sAction[0:1] == "=":
            sAction = sAction[1:]
        lFUNCTIONS.append(("_d_"+sActionId, sAction))
        sAction = "_d_"+sActionId
        return [sCondition, cAction, sAction]
    print(f"# Unknown action at line {sLineId} / {sActionId}")
    return None


def _calcRulesStats (lRules):
    "count rules and actions"
    d = {'=':0, '~': 0, '-': 0, '>': 0}
    for aRule in lRules: