json-repair 0.9.0__py3-none-any.whl → 0.10.1__py3-none-any.whl

json_repair/json_repair.py

@@ -61,7 +61,9 @@ class JSONParser:
         elif char == '"':
             return self.parse_string()
         elif char == "'":
-            return self.parse_string(use_single_quotes=True)
+            return self.parse_string(string_quotes="'")
+        elif char == "“":
+            return self.parse_string(string_quotes=["“", "”"])
         # <number> starts with [0-9] or minus
         elif char.isdigit() or char == "-":
             return self.parse_number()
@@ -102,9 +104,7 @@ class JSONParser:
             # <member> starts with a <string>
             key = ""
             while key == "" and self.get_char_at():
-                key = self.parse_string(
-                    use_single_quotes=(self.json_str[self.index] == "'")
-                )
+                key = self.parse_json()
 
                 # This can happen sometimes like { "": "value" }
                 if key == "" and self.get_char_at() == ":":
@@ -112,7 +112,7 @@ class JSONParser:
                     break
 
             # We reached the end here
-            if key == "}":
+            if (self.get_char_at() or "}") == "}":
                 continue
 
             # An extreme case of missing ":" after a key
@@ -170,19 +170,22 @@ class JSONParser:
         self.index += 1
         return arr
 
-    def parse_string(self, use_single_quotes=False) -> str:
+    def parse_string(self, string_quotes=False) -> str:
         # <string> is a string of valid characters enclosed in quotes
         # i.e. { name: "John" }
         # Somehow all weird cases in an invalid JSON happen to be resolved in this function, so be careful here
 
         # Flag to manage corner cases related to missing starting quote
         fixed_quotes = False
-        string_terminator = '"'
-        if use_single_quotes:
-            string_terminator = "'"
+        lstring_delimiter = rstring_delimiter = '"'
+        if isinstance(string_quotes, list):
+            lstring_delimiter = string_quotes[0]
+            rstring_delimiter = string_quotes[1]
+        elif isinstance(string_quotes, str):
+            lstring_delimiter = rstring_delimiter = string_quotes
         char = self.get_char_at()
-        if char != string_terminator:
-            self.insert_char_at(string_terminator)
+        if char != lstring_delimiter:
+            self.insert_char_at(lstring_delimiter)
             fixed_quotes = True
         else:
             self.index += 1
@@ -198,7 +201,7 @@ class JSONParser:
         # * If we are fixing missing quotes in an object, when it finds the special terminators
         char = self.get_char_at()
         fix_broken_markdown_link = False
-        while char and char != string_terminator:
+        while char and char != rstring_delimiter:
             if fixed_quotes:
                 if self.context == "object_key" and (char == ":" or char.isspace()):
                     break
@@ -206,9 +209,13 @@ class JSONParser:
                     break
             self.index += 1
             char = self.get_char_at()
+            # If the string contains escaped delimiters we should respect that
+            if char == rstring_delimiter and self.get_char_at(-1) == "\\":
+                self.index += 1
+                char = self.get_char_at()
             # ChatGPT sometimes forget to quote links in markdown like: { "content": "[LINK]("https://google.com")" }
             if (
-                char == string_terminator
+                char == rstring_delimiter
                 # Next character is not a comma
                 and self.get_char_at(1) != ","
                 and (
@@ -228,8 +235,8 @@ class JSONParser:
         end = self.index
 
         # A fallout of the previous special case in the while loop, we need to update the index only if we had a closing quote
-        if char != string_terminator:
-            self.insert_char_at(string_terminator)
+        if char != rstring_delimiter:
+            self.insert_char_at(rstring_delimiter)
         else:
             self.index += 1
 

{json_repair-0.9.0.dist-info → json_repair-0.10.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: json_repair
-Version: 0.9.0
+Version: 0.10.1
 Summary: A package to repair broken json strings
 Author-email: Stefano Baccianella <4247706+mangiucugna@users.noreply.github.com>
 License: MIT License
@@ -56,20 +56,20 @@ I searched for a lightweight python package that was able to reliably fix this p
 You can use this library to completely replace `json.loads()`:
 
     import json_repair
-    
+
     decoded_object = json_repair.loads(json_string)
 
 or just
 
     import json_repair
-    
+
     decoded_object = json_repair.repair_json(json_string, return_objects=True)
-    
+
 ### Performance considerations
 If you find this library too slow because is using `json.loads()` you can skip that by passing `skip_json_loads=True` to `repair_json`. Like:
 
     from json_repair import repair_json
-    
+
     good_json_string = repair_json(bad_json_string, skip_json_loads=True)
 
 I made a choice of not using any fast json library to avoid having any external dependency, so that anybody can use it regardless of their stack.
@@ -77,7 +77,7 @@ I made a choice of not using any fast json library to avoid having any external
 Some rules of thumb to use:
 - Setting `return_objects=True` will always be faster because the parser returns an object already and it doesn't have serialize that object to JSON
 - `skip_json_loads` is faster only if you 100% know that the string is not a valid JSON
-
+- If you are having issues with escaping pass the string as **raw** string like: `r"string with escaping\""`
 ## Adding to requirements
 **Please pin this library only on the major version!**
 

json_repair-0.10.1.dist-info/RECORD

@@ -0,0 +1,7 @@
+json_repair/__init__.py,sha256=p9mZnte8Bg18NcxqgJ7vopH2gQv_XbZ0dRnk686QuRE,92
+json_repair/json_repair.py,sha256=OIjr9L0CyysGrEEdfeeEkKoiFLaX2sGSnn2MYBqGHHo,13826
+json_repair-0.10.1.dist-info/LICENSE,sha256=wrjQo8MhNrNCicXtMe3MHmS-fx8AmQk1ue8AQwiiFV8,1076
+json_repair-0.10.1.dist-info/METADATA,sha256=kPI-1mCI6HuVeV0eVV98GR5BS6Tk04tkxDkxKCdKG7E,6404
+json_repair-0.10.1.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
+json_repair-0.10.1.dist-info/top_level.txt,sha256=7-VZwZN2CgB_n0NlSLk-rEUFh8ug21lESbsblOYuZqw,12
+json_repair-0.10.1.dist-info/RECORD,,

json_repair-0.9.0.dist-info/RECORD

@@ -1,7 +0,0 @@
-json_repair/__init__.py,sha256=p9mZnte8Bg18NcxqgJ7vopH2gQv_XbZ0dRnk686QuRE,92
-json_repair/json_repair.py,sha256=63d6OrbAW3dbJ0C06eGtxSB8x5EASRI3OvdsXvmotKk,13373
-json_repair-0.9.0.dist-info/LICENSE,sha256=wrjQo8MhNrNCicXtMe3MHmS-fx8AmQk1ue8AQwiiFV8,1076
-json_repair-0.9.0.dist-info/METADATA,sha256=ueRJPKZhwXF66UmZXxcLNVwkOJbjGSbsRR8hhKo78MI,6311
-json_repair-0.9.0.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
-json_repair-0.9.0.dist-info/top_level.txt,sha256=7-VZwZN2CgB_n0NlSLk-rEUFh8ug21lESbsblOYuZqw,12
-json_repair-0.9.0.dist-info/RECORD,,