Grammalecte  Check-in [b5ee5c589f]

Overview
Comment:[build] check token validity
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk | build
Files: files | file ages | folders
SHA3-256: b5ee5c589f7db37389be4f363c770ea4eb848e127f605f60312246203cfa7941
User & Date: olr on 2020-11-23 10:32:13
Other Links: manifest | tags
Context
2020-11-23
20:28
[fr] corrections des messages et ajustements check-in: ab708276ae user: olr tags: trunk, fr
10:32
[build] check token validity check-in: b5ee5c589f user: olr tags: trunk, build
2020-11-21
21:42
[core][fr] tests update check-in: 6639af3543 user: olr tags: trunk, fr, core
Changes

Modified compile_rules_graph.py from [38bc0eb9da] to [dfe32ac11e].

103
104
105
106
107
108
109



























110
111
112


113
114
115
116
117
118
119
120



121


122
123
124
125
126
127
128
        self.dOptPriority = dOptPriority
        self.dAntiPatterns = {}
        self.dActions = {}
        self.dFuncName = {}
        self.dFunctions = {}
        self.dLemmas = {}




























    def _genTokenLines (self, sTokenLine):
        "tokenize a string and return a list of lines of tokens"
        lTokenLines = []


        for sTokBlock in sTokenLine.split():
            # replace merger characters by spaces
            if "␣" in sTokBlock:
                sTokBlock = sTokBlock.replace("␣", " ")
            # optional token?
            bNullPossible = sTokBlock.startswith("?") and sTokBlock.endswith("¿")
            if bNullPossible:
                sTokBlock = sTokBlock[1:-1]



            # token with definition?


            if sTokBlock.startswith("({") and sTokBlock.endswith("})") and sTokBlock[1:-1] in self.dDef:
                sTokBlock = "(" + self.dDef[sTokBlock[1:-1]] + ")"
            elif sTokBlock.startswith("{") and sTokBlock.endswith("}") and sTokBlock in self.dDef:
                sTokBlock = self.dDef[sTokBlock]
            if ( (sTokBlock.startswith("[") and sTokBlock.endswith("]")) or (sTokBlock.startswith("([") and sTokBlock.endswith("])")) ):
                # multiple token
                bSelectedGroup = sTokBlock.startswith("(") and sTokBlock.endswith(")")







>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>



>
>
|







>
>
>

>
>







103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
        self.dOptPriority = dOptPriority
        self.dAntiPatterns = {}
        self.dActions = {}
        self.dFuncName = {}
        self.dFunctions = {}
        self.dLemmas = {}

    def createGraphAndActions (self, lRuleLine):
        "create a graph as a dictionary with <lRuleLine>"
        fStartTimer = time.time()
        print("{:>8,} rules in {:<30} ".format(len(lRuleLine), f"<{self.sGraphName}|{self.sGraphCode}>"), end="")
        lPreparedRule = []
        for i, sRuleName, sTokenLine, iActionBlock, lActions, nPriority in lRuleLine:
            for aRule in self.createRule(i, sRuleName, sTokenLine, iActionBlock, lActions, nPriority):
                lPreparedRule.append(aRule)
        # Debugging
        if False:
            print("\nRULES:")
            for e in lPreparedRule:
                if e[-2] == "##2211":
                    print(e)
        # Graph creation
        oDARG = darg.DARG(lPreparedRule, self.sLang)
        dGraph = oDARG.createGraph()
        print(oDARG, end="")
        # debugging
        if False:
            print("\nGRAPH:", self.sGraphName)
            for k, v in dGraph.items():
                print(k, "\t", v)
        print("\tin {:>8.2f} s".format(time.time()-fStartTimer))
        sPyCallables, sJSCallables = self.createCallables()
        return dGraph, self.dActions, sPyCallables, sJSCallables, self.dLemmas

    def _genTokenLines (self, sTokenLine):
        "tokenize a string and return a list of lines of tokens"
        lTokenLines = []
        nFirstNullable = 0
        nLastNullable = 0
        for n, sTokBlock in enumerate(sTokenLine.split(), 1):
            # replace merger characters by spaces
            if "␣" in sTokBlock:
                sTokBlock = sTokBlock.replace("␣", " ")
            # optional token?
            bNullPossible = sTokBlock.startswith("?") and sTokBlock.endswith("¿")
            if bNullPossible:
                sTokBlock = sTokBlock[1:-1]
                if nFirstNullable == 0:
                    nFirstNullable = n
                nLastNullable = n
            # token with definition?
            if sTokBlock.startswith("(") and sTokBlock.endswith(")"):
                nFirstNullable = -1
            if sTokBlock.startswith("({") and sTokBlock.endswith("})") and sTokBlock[1:-1] in self.dDef:
                sTokBlock = "(" + self.dDef[sTokBlock[1:-1]] + ")"
            elif sTokBlock.startswith("{") and sTokBlock.endswith("}") and sTokBlock in self.dDef:
                sTokBlock = self.dDef[sTokBlock]
            if ( (sTokBlock.startswith("[") and sTokBlock.endswith("]")) or (sTokBlock.startswith("([") and sTokBlock.endswith("])")) ):
                # multiple token
                bSelectedGroup = sTokBlock.startswith("(") and sTokBlock.endswith(")")
161
162
163
164
165
166
167

168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
                            lNew = list(aRule)
                            lNew.append(sTokBlock)
                            lNewTemp.append(lNew)
                        lTokenLines.extend(lNewTemp)
                    else:
                        for aRule in lTokenLines:
                            aRule.append(sTokBlock)

        for aRule in lTokenLines:
            yield aRule

    def _createTokenList (self, sTokBlock):
        "return a list of tokens from a block of tokens"
        lToken = []
        for sToken in sTokBlock[1:-1].split("|"):
            if "+" in sToken and not sToken.startswith("+"):
                for sCode in self.dDecl:
                    if sToken.endswith(sCode):
                        sToken = sToken[:-len(sCode)]
                        lToken.append(sToken)
                        for sSuffix in self.dDecl[sCode]:
                            lToken.append(sToken+sSuffix)
                        break
            else:
                lToken.append(sToken)
        return lToken

    def createGraphAndActions (self, lRuleLine):
        "create a graph as a dictionary with <lRuleLine>"
        fStartTimer = time.time()
        print("{:>8,} rules in {:<30} ".format(len(lRuleLine), f"<{self.sGraphName}|{self.sGraphCode}>"), end="")
        lPreparedRule = []
        for i, sRuleName, sTokenLine, iActionBlock, lActions, nPriority in lRuleLine:
            for aRule in self.createRule(i, sRuleName, sTokenLine, iActionBlock, lActions, nPriority):
                lPreparedRule.append(aRule)
        # Debugging
        if False:
            print("\nRULES:")
            for e in lPreparedRule:
                if e[-2] == "##2211":
                    print(e)
        # Graph creation
        oDARG = darg.DARG(lPreparedRule, self.sLang)
        dGraph = oDARG.createGraph()
        print(oDARG, end="")
        # debugging
        if False:
            print("\nGRAPH:", self.sGraphName)
            for k, v in dGraph.items():
                print(k, "\t", v)
        print("\tin {:>8.2f} s".format(time.time()-fStartTimer))
        sPyCallables, sJSCallables = self.createCallables()
        return dGraph, self.dActions, sPyCallables, sJSCallables, self.dLemmas

    def createRule (self, iLine, sRuleName, sTokenLine, iActionBlock, lActions, nPriority):
        "generator: create rule as list"
        # print(iLine, "//", sRuleName, "//", sTokenLine, "//", lActions, "//", nPriority)
        if sTokenLine.startswith("!!") and sTokenLine.endswith("¡¡"):
            # antipattern
            sTokenLine = sTokenLine[2:-2].strip()
            if sRuleName not in self.dAntiPatterns:
                self.dAntiPatterns[sRuleName]= []
            for lToken in self._genTokenLines(sTokenLine):
                self.dAntiPatterns[sRuleName].append(lToken)
        else:
            # pattern
            for lToken in self._genTokenLines(sTokenLine):
                if sRuleName in self.dAntiPatterns and lToken in self.dAntiPatterns[sRuleName]:
                    # <lToken> matches an antipattern -> discard
                    continue
                # Calculate positions
                dPos = {}   # key: iGroup, value: iToken
                iGroup = 0
                #if iLine == 15818: # debug







>

|

















<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<








|



|







195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221



























222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
                            lNew = list(aRule)
                            lNew.append(sTokBlock)
                            lNewTemp.append(lNew)
                        lTokenLines.extend(lNewTemp)
                    else:
                        for aRule in lTokenLines:
                            aRule.append(sTokBlock)
        nLastNullable = nLastNullable - n - 1
        for aRule in lTokenLines:
            yield aRule, nFirstNullable, nLastNullable

    def _createTokenList (self, sTokBlock):
        "return a list of tokens from a block of tokens"
        lToken = []
        for sToken in sTokBlock[1:-1].split("|"):
            if "+" in sToken and not sToken.startswith("+"):
                for sCode in self.dDecl:
                    if sToken.endswith(sCode):
                        sToken = sToken[:-len(sCode)]
                        lToken.append(sToken)
                        for sSuffix in self.dDecl[sCode]:
                            lToken.append(sToken+sSuffix)
                        break
            else:
                lToken.append(sToken)
        return lToken




























    def createRule (self, iLine, sRuleName, sTokenLine, iActionBlock, lActions, nPriority):
        "generator: create rule as list"
        # print(iLine, "//", sRuleName, "//", sTokenLine, "//", lActions, "//", nPriority)
        if sTokenLine.startswith("!!") and sTokenLine.endswith("¡¡"):
            # antipattern
            sTokenLine = sTokenLine[2:-2].strip()
            if sRuleName not in self.dAntiPatterns:
                self.dAntiPatterns[sRuleName]= []
            for lToken, _, _ in self._genTokenLines(sTokenLine):
                self.dAntiPatterns[sRuleName].append(lToken)
        else:
            # pattern
            for lToken, nFirstNullable, nLastNullable in self._genTokenLines(sTokenLine):
                if sRuleName in self.dAntiPatterns and lToken in self.dAntiPatterns[sRuleName]:
                    # <lToken> matches an antipattern -> discard
                    continue
                # Calculate positions
                dPos = {}   # key: iGroup, value: iToken
                iGroup = 0
                #if iLine == 15818: # debug
241
242
243
244
245
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
                    if sToken.startswith(">") and sToken != ">" and sToken[1:] not in self.dLemmas:
                        self.dLemmas[sToken[1:]] = iLine
                # Parse actions
                for iAction, (iActionLine, sAction) in enumerate(lActions, 1):
                    sAction = sAction.strip()
                    if sAction:
                        sActionId = f"{self.sGraphCode}__{sRuleName}__b{iActionBlock}_a{iAction}"
                        aAction = self.createAction(sActionId, sAction, nPriority, len(lToken), dPos, iActionLine)
                        if aAction:
                            sActionName = self.storeAction(sActionId, aAction)
                            lResult = list(lToken)
                            lResult.extend(["##"+str(iLine), sActionName])
                            #if iLine == 13341:
                            #    print("  ".join(lToken))
                            #    print(sActionId, aAction)
                            yield lResult
                        else:
                            print("# Error on action at line:", iLine)
                            print(sTokenLine, "\n", lActions)
                            exit()
                    else:
                        print("No action found for ", iActionLine)
                        exit()

    def createAction (self, sActionId, sAction, nPriority, nToken, dPos, iActionLine):
        "create action rule as a list"
        sLineId = "#" + str(iActionLine)

        # Option
        sOption = False
        m = re.match("/(\\w+)/", sAction)
        if m:







|
















|







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
                    if sToken.startswith(">") and sToken != ">" and sToken[1:] not in self.dLemmas:
                        self.dLemmas[sToken[1:]] = iLine
                # Parse actions
                for iAction, (iActionLine, sAction) in enumerate(lActions, 1):
                    sAction = sAction.strip()
                    if sAction:
                        sActionId = f"{self.sGraphCode}__{sRuleName}__b{iActionBlock}_a{iAction}"
                        aAction = self.createAction(sActionId, sAction, nPriority, len(lToken), dPos, iActionLine, nFirstNullable, nLastNullable)
                        if aAction:
                            sActionName = self.storeAction(sActionId, aAction)
                            lResult = list(lToken)
                            lResult.extend(["##"+str(iLine), sActionName])
                            #if iLine == 13341:
                            #    print("  ".join(lToken))
                            #    print(sActionId, aAction)
                            yield lResult
                        else:
                            print("# Error on action at line:", iLine)
                            print(sTokenLine, "\n", lActions)
                            exit()
                    else:
                        print("No action found for ", iActionLine)
                        exit()

    def createAction (self, sActionId, sAction, nPriority, nToken, dPos, iActionLine, nFirstNullable, nLastNullable):
        "create action rule as a list"
        sLineId = "#" + str(iActionLine)

        # Option
        sOption = False
        m = re.match("/(\\w+)/", sAction)
        if m:
292
293
294
295
296
297
298

299
300
301
302
303
304
305
        # Case sensitivity
        bCaseSensitivity = not bool(m.group("casing"))

        # Action
        cAction = m.group("action")
        sAction = sAction[m.end():].strip()
        sAction = changeReferenceToken(sAction, dPos)

        # target
        cStartLimit = "<"
        cEndLimit = ">"
        if not m.group("start"):
            iStartAction = 1
            iEndAction = 0
        else:







>







300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
        # Case sensitivity
        bCaseSensitivity = not bool(m.group("casing"))

        # Action
        cAction = m.group("action")
        sAction = sAction[m.end():].strip()
        sAction = changeReferenceToken(sAction, dPos)

        # target
        cStartLimit = "<"
        cEndLimit = ">"
        if not m.group("start"):
            iStartAction = 1
            iEndAction = 0
        else:
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
            if iEndAction:
                iEndAction = dPos.get(iEndAction, iEndAction)
        if iStartAction < 0:
            iStartAction += 1
        if iEndAction < 0:
            iEndAction += 1












        if cAction == "-":
            ## error
            iMsg = sAction.find(" && ")
            if iMsg == -1:
                sMsg = "# Error. Error message not found."
                sURL = ""
                print("\n# Error. No message at: ", sLineId, sActionId)
                exit()
            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()
                checkTokenNumbers(sMsg, sActionId, nToken)
                if sMsg[0:1] == "=":
                    sMsg = self.createFunction("msg", sMsg, True)
                else:
                    checkIfThereIsCode(sMsg, sActionId)

        # checking consistancy

        checkTokenNumbers(sAction, sActionId, nToken)

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

        if not sAction and cAction != "!":
            print(f"\n# Error in action at line <{sLineId}/{sActionId}>:  This action is empty.")







>
>
>
>
>
>
>
>
>
>
>




<
<










|





|
>
|







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
376
            if iEndAction:
                iEndAction = dPos.get(iEndAction, iEndAction)
        if iStartAction < 0:
            iStartAction += 1
        if iEndAction < 0:
            iEndAction += 1

        # check target
        if nFirstNullable > -1:
            if nFirstNullable > 0 and iStartAction > 0 and iEndAction != 0 and iStartAction > nFirstNullable:
                print(f"# Error. At {sLineId}, {sActionId}, target start is bigger than first nullable token.")
            if nFirstNullable > 0 and iEndAction > 0 and iStartAction != 1 and iEndAction > nFirstNullable:
                print(f"# Error. At {sLineId}, {sActionId}, target end is bigger than first nullable token.")
            if nLastNullable < 0 and iStartAction < 0 and iEndAction != 0 and (iStartAction-1) < nLastNullable:
                print(f"# Error. At {sLineId}, {sActionId}, target start is lower than last nullable token.")
            if nLastNullable < 0 and iEndAction < 0 and iStartAction != 1 and (iEndAction-1) < nLastNullable:
                print(f"# Error. At {sLineId}, {sActionId}, target end is lower than last nullable token.")

        if cAction == "-":
            ## error
            iMsg = sAction.find(" && ")
            if iMsg == -1:


                print("\n# Error. No message at: ", sLineId, sActionId)
                exit()
            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()
                checkTokenNumbers(sMsg, sActionId, nToken)      # check tokens in message
                if sMsg[0:1] == "=":
                    sMsg = self.createFunction("msg", sMsg, True)
                else:
                    checkIfThereIsCode(sMsg, sActionId)

        # checking token consistancy
        checkTokenNumbers(sCondition, sActionId, nToken)    # check tokens in condition
        checkTokenNumbers(sAction, sActionId, nToken)       # check tokens in action

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

        if not sAction and cAction != "!":
            print(f"\n# Error in action at line <{sLineId}/{sActionId}>:  This action is empty.")