update python script

.gitignore

@@ -9,6 +9,7 @@ NlpParser/*
 !NlpParser/NlpRunner.java
 !NlpParser/testbench.txt
 !NlpParser/README.md
+NlpTr/*
 
 .vscode/
 

NlpProc/NlpDirectCmp.py

@@ -0,0 +1,12 @@
+import NlpUtils
+import jsondiff
+import sys
+
+json1 = NlpUtils.LoadJsonFromFile(sys.argv[1])
+json2 = NlpUtils.LoadJsonFromFile(sys.argv[2])
+
+(_, value1, ) = NlpUtils.NlpJson2PlainJsonWrapper(json1)
+(_, value2, ) = NlpUtils.NlpJson2PlainJsonWrapper(json2)
+
+diff = jsondiff.diff(value1, value2)
+print(diff)

NlpProc/NlpJsonDecoder.py

@@ -0,0 +1,55 @@
+import NlpUtils
+import jsondiff
+import sys, collections
+
+CmdArgvPair = collections.namedtuple('CmdArgvPair', ('nlpJson', 'trTemplate', 'trDiff', 'trIndex'))
+def CmdArgvAnalyzer() -> tuple[CmdArgvPair]:
+    ls: list[CmdArgvPair] = []
+
+    argc = len(sys.argv) - 1
+    if argc % 4 != 0:
+        print("invalid parameter.")
+        sys.exit(1)
+
+    count = argc // 4
+    return tuple(CmdArgvPair._make(sys.argv[1 + i * 4:5 + i * 4]) for i in range(count))
+
+# script will order 1 file as reference
+# 0. the nlp json file
+# script will output 3 files for each version translation.
+# 0. translation template json
+# 0. the diff result comparing with the previous version
+# 0. a list including the key of each value in template json
+# so for a single version virtools, we need input 4 arguments
+if __name__ == "__main__":
+    resolvedArgv = CmdArgvAnalyzer()
+
+    prevJson = None
+    for vtVer in resolvedArgv:
+        # read nlp json and convert it into plain json
+        nlpJson = NlpUtils.LoadJson(vtVer.nlpJson)
+        (plainKeys, plainValues, ) = NlpUtils.NlpJson2PlainJsonWrapper(nlpJson)
+
+        # write index file
+        NlpUtils.DumpTrIndex(vtVer.trIndex, plainKeys)
+
+        # compare with previous one
+        if prevJson is None:
+            # this is first json. omit diff
+            # write blank diff and write whole translation values
+            NlpUtils.DumpTrDiff(vtVer.trDiff, [], [])
+            NlpUtils.DumpTrTemplate(vtVer.trTemplate, dict(zip(plainKeys, plainValues)))
+        else:
+            # compare with prev json
+            cmpResult = jsondiff.diff(prevJson, plainValues)
+            # seperate diff result
+            (insertedKey, deletedKey, insertedVal) = NlpUtils.SeperatePlainJsonDiff(cmpResult)
+
+            # write diff
+            NlpUtils.DumpTrDiff(vtVer.trDiff, insertedKey, deletedKey)
+            # write template with special treat
+            NlpUtils.DumpTrTemplate(vtVer.trTemplate, dict((i, j) for i, j in enumerate(insertedVal)))
+
+        # assign prev json
+        prevJson = plainValues
+

NlpProc/NlpTrBaseCreator.py

@@ -1,26 +0,0 @@
-import NlpUtils
-import sys, collections
-
-def CreateBaseJsonWrapper(baseJson: dict) -> dict[str, str]:
-    result: dict[str, str] = {}
-    stack: collections.deque = collections.deque()
-    CreateBaseJson(baseJson, stack, result)
-    return result
-def CreateBaseJson(baseJson: dict, stack: collections.deque, result: dict[str, str]):
-    assert isinstance(baseJson, dict)
-    assert 'entries' in baseJson
-
-    counter = 0
-    for entry in baseJson['entries']:
-        if isinstance(entry, str):
-            result['.'.join(tuple(stack) + (str(counter), ))] = entry
-            counter += 1
-        else:
-            stack.append(entry['section'])
-            CreateBaseJson(entry, stack, result)
-            stack.pop()
-
-if __name__ == "__main__":
-    baseJson = NlpUtils.LoadJsonFromFile(sys.argv[1])
-    trJson = CreateBaseJsonWrapper(baseJson)
-    NlpUtils.WriteJsonToFile(sys.argv[2], trJson)

NlpProc/NlpUtils.py

@@ -3,44 +3,100 @@ import collections
 import io
 import json
 
-def WriteJsonToFile(filepath: str, jsonData: dict):
+def DumpJson(filepath: str, jsonData: dict):
     with open(filepath, 'w', encoding='utf-8') as f:
         json.dump(jsonData, f, indent=4, sort_keys=False)
 
-def LoadJsonFromFile(filepath: str) -> dict:
+def LoadJson(filepath: str) -> dict:
     with open(filepath, 'r', encoding='utf-8') as f:
         return json.load(f)
 
-def SaveDiffToFileWrapper(jsonDiffData: dict, filepath: str) -> dict[str, str]:
-    result: dict[tuple[int], str] = {}
-    stack: collections.deque = collections.deque()
-    with open(filepath, 'w', encoding='utf-8') as fdiff:
-        SaveDiffToFile(jsonDiffData, fdiff, stack, result)
-    return result
+def DumpTrIndex(filepath: str, indexData: list[str]):
+    with open(filepath, 'w', encoding='utf-8') as f:
+        for item in indexData:
+            f.write(item)
+            f.write('\n')
 
-def SaveDiffToFile(jsonDiffData: dict, fs: io.TextIOWrapper, stack: collections.deque, result: dict[tuple[int], str]):
-    assert isinstance(jsonDiffData, dict)
-    assert len(jsonDiffData) == 1
-    assert "entries" in jsonDiffData
-    assert isinstance(jsonDiffData["entries"], dict)
+def LoadTrIndex(filepath: str) -> list[str]:
+    data: list[str] = []
+    with open(filepath, 'r', encoding='utf-8') as f:
+        while True:
+            ln = f.readline()
+            if ln == '': break
+            data.append(ln.strip('\n'))
 
-    for key, item in jsonDiffData["entries"].items():
-        if isinstance(key, int):
-            stack.append(key)
-            SaveDiffToFile(item, fs, stack, result)
-            stack.pop()
-        elif key == jsondiff.symbols.insert:
-            for (modIdx, modEntry) in item:
-                stridx = ".".join(tuple(stack) + (modIdx, ))
-                result[stridx] = modEntry
-                fs.write(f'i {stridx}\n')
-        elif key == jsondiff.symbols.delete:
-            for delIdx in item:
-                stridx = ".".join(tuple(stack) + (delIdx, ))
-                fs.write(f'd {stridx}\n')
+    return data
+
+def DumpTrTemplate(filepath: str, templateData: dict[str, str]):
+    DumpJson(filepath, templateData)
+
+def LoadTrTemplate(filepath: str) ->  dict[str, str]:
+    return LoadJson(filepath)
+
+def DumpTrDiff(filepath: str, insertedKey: list[str], deletedKey: list[str]):
+    with open(filepath, 'w', encoding='utf-8') as f:
+        for entryIdx in insertedKey:
+            f.write(f'i/{entryIdx}\n')
+
+        for entryIdx in deletedKey:
+            f.write(f'd/{entryIdx}\n')
+
+# return a tuple. (insertedKey, deletedKey)
+def LoadTrDiff(filepath: str) -> dict:
+    insertedKey: list[str] = []
+    deletedKey: list[str] = []
+    with open(filepath, 'r', encoding='utf-8') as f:
+        while True:
+            ln = f.readline()
+            if ln == '': break
+
+            sp = ln.strip('\n').split('/')
+            if sp[0] == 'i':
+                insertedKey.append(sp[1])
             else:
-            raise Exception("invalid key type")
+                deletedKey.append(sp[1])
 
-def ReadDiffFromFile(translations: tuple[str], filepath: str, result: dict):
-    pass
+    return (insertedKey, deletedKey)
+
+# return a tuple. (insertedKey, deletedKey, insertedVal)
+def SeperatePlainJsonDiff(diffData: dict) -> tuple:
+    insertedKey: list[str] = []
+    insertedVal: list[str] = []
+
+    if jsondiff.insert in diffData:
+        for (entryIdx, entryVal, ) in diffData[jsondiff.insert]:
+            insertedKey.append(entryIdx)
+            insertedVal.append(entryVal)
+
+    if jsondiff.delete in diffData:
+        deletedKey = diffData[jsondiff.delete][:]
+    else:
+        deletedKey = []
+
+    return (insertedKey, deletedKey, insertedVal)
+
+# return a tuple. (keyList, valueList)
+def NlpJson2PlainJsonWrapper(nlpJson: dict) -> tuple:
+    keyList: list[str] = []
+    valueList: list[str] = []
+    stack: collections.deque = collections.deque()
+    NlpJson2PlainJson(nlpJson, stack, keyList, valueList)
+    return (keyList, valueList, )
+def NlpJson2PlainJson(nlpJson: dict, stack: collections.deque, keyList: list[str], valueList: list[str]):
+    assert isinstance(nlpJson, dict)
+    assert 'entries' in nlpJson
+
+    counter = 0
+    for entry in nlpJson['entries']:
+        if isinstance(entry, str):
+            # is data node. add into result
+            keyList.append('/'.join(tuple(stack) + (str(counter), )))
+            valueList.append(entry)
+            counter += 1
+        else:
+            # is a sub section
+            # push section name and recursive calling this function
+            stack.append(entry['section'])
+            NlpJson2PlainJson(entry, stack, keyList, valueList)
+            stack.pop()
 

NlpProc/README.md

@@ -0,0 +1,13 @@
+# Nlp Proc
+
+Example:
+
+`py NlpJsonDecoder.py ../NlpSrc/VT25.json ../NlpTr/VT25.template.json ../NlpTr/VT25.diff ../NlpTr/VT25.index ../NlpSrc/VT35.json ../NlpTr/VT35.template.json ../NlpTr/VT35.diff ../NlpTr/VT35.index ../NlpSrc/VT40.json ../NlpTr/VT40.template.json ../NlpTr/VT40.diff ../NlpTr/VT40.index ../NlpSrc/VT50.json ../NlpTr/VT50.template.json ../NlpTr/VT50.diff ../NlpTr/VT50.index`
+
+```
+py NlpJsonDecoder.py
+../NlpSrc/VT25.json ../NlpTr/VT25.template.json ../NlpTr/VT25.diff ../NlpTr/VT25.index
+../NlpSrc/VT35.json ../NlpTr/VT35.template.json ../NlpTr/VT35.diff ../NlpTr/VT35.index
+../NlpSrc/VT40.json ../NlpTr/VT40.template.json ../NlpTr/VT40.diff ../NlpTr/VT40.index
+../NlpSrc/VT50.json ../NlpTr/VT50.template.json ../NlpTr/VT50.diff ../NlpTr/VT50.index
+```