json-repair 0.9.0__tar.gz → 0.10.1__tar.gz

{json_repair-0.9.0/src/json_repair.egg-info → json_repair-0.10.1}/PKG-INFO

@@ -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-0.9.0 → json_repair-0.10.1}/README.md

@@ -19,20 +19,20 @@ 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.
@@ -40,7 +40,7 @@ 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-0.9.0 → json_repair-0.10.1}/pyproject.toml

@@ -3,7 +3,7 @@ requires = ["setuptools>=61.0"]
 build-backend = "setuptools.build_meta"
 [project]
 name = "json_repair"
-version = "0.9.0"
+version = "0.10.1"
 license = {file = "LICENSE"}
 authors = [
   { name="Stefano Baccianella", email="4247706+mangiucugna@users.noreply.github.com" },

{json_repair-0.9.0 → json_repair-0.10.1}/src/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 → json_repair-0.10.1/src/json_repair.egg-info}/PKG-INFO

@@ -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-0.9.0 → json_repair-0.10.1}/tests/test_json_repair.py

@@ -50,12 +50,13 @@ def test_repair_json():
     assert repair_json("[[1\n\n]") == "[[1]]"
     assert repair_json("{") == "{}"
     assert repair_json("}") == '""'
-    assert repair_json('{"') == '{"": ""}'
+    assert repair_json('{"') == '{}'
     assert repair_json('["') == '[]'
     assert repair_json("'\"'") == '"\\\""'
     assert repair_json("'string\"") == '"string\\\""'
     assert repair_json('{foo: [}') == '{"foo": []}'
     assert repair_json('{"key": "value:value"}') == '{"key": "value:value"}'
+    assert repair_json('{“slanted_delimiter”: "value"}') == '{"slanted_delimiter": "value"}'
     assert (
         repair_json('{"name": "John", "age": 30, "city": "New')
         == '{"name": "John", "age": 30, "city": "New"}'
@@ -102,6 +103,7 @@ def test_repair_json():
                        { "key": "value" }
                        ```""") == '{"key": "value"}'
     assert repair_json('````{ "key": "value" }```') == '{"key": "value"}'
+    assert repair_json(r'{"real_content": "Some string: Some other string Some string <a href=\"https://domain.com\">Some link</a>"') == r'{"real_content": "Some string: Some other string Some string <a href=\\\"https://domain.com\\\">Some link</a>"}'
 
 
 

{json_repair-0.9.0 → json_repair-0.10.1}/tests/test_performance.py

@@ -580,7 +580,7 @@ def test_true_false_correct(benchmark):
   mean_time = benchmark.stats.get("median")
 
   # Define your time threshold in seconds
-  max_time = 22 * (1 / 10 ** 6)  # 21 microsecond
+  max_time = 23 * (1 / 10 ** 6)  # 23 microsecond
 
   # Assert that the average time is below the threshold
   assert mean_time < max_time, f"Benchmark exceeded threshold: {mean_time:.3f}s > {max_time:.3f}s"
@@ -624,7 +624,7 @@ def test_false_false_correct(benchmark):
   mean_time = benchmark.stats.get("median")
 
   # Define your time threshold in seconds
-  max_time = 54 / 10 ** 6  # 54 microsecond
+  max_time = 56 / 10 ** 6  # 56 microsecond
 
   # Assert that the average time is below the threshold
   assert mean_time < max_time, f"Benchmark exceeded threshold: {mean_time:.3f}s > {max_time:.3f}s"