json-repair 0.23.1__py3-none-any.whl → 0.25.0__py3-none-any.whl

json_repair/json_repair.py

@@ -91,6 +91,10 @@ class JSONParser:
     ) -> Union[JSONReturnType, Tuple[JSONReturnType, List[Dict[str, str]]]]:
         json = self.parse_json()
         if self.index < len(self.json_str):
+            self.log(
+                "The parser returned early, checking if there's more json elements",
+                "info",
+            )
             json = [json]
             last_index = self.index
             while self.index < len(self.json_str):
@@ -100,10 +104,13 @@ class JSONParser:
                 if self.index == last_index:
                     self.index += 1
                 last_index = self.index
+            # If nothing extra was found, don't return an array
             if len(json) == 1:
+                self.log(
+                    "There were no more elements, returning the element without the array",
+                    "info",
+                )
                 json = json[0]
-            elif len(json) == 0:
-                json = ""
         if self.logger.log_level == "none":
             return json
         else:
@@ -363,9 +370,34 @@ class JSONParser:
                 if self.get_context() == "object_key" and (
                     char == ":" or char.isspace()
                 ):
+                    self.log(
+                        "While parsing a string missing the left delimiter in object key context, we found a :, stopping here",
+                        "info",
+                    )
                     break
                 elif self.get_context() == "object_value" and char in [",", "}"]:
-                    break
+                    rstring_delimiter_missing = True
+                    # check if this is a case in which the closing comma is NOT missing instead
+                    i = 1
+                    next_c = self.get_char_at(i)
+                    while next_c and next_c != rstring_delimiter:
+                        i += 1
+                        next_c = self.get_char_at(i)
+                    if next_c:
+                        i += 1
+                        next_c = self.get_char_at(i)
+                        # found a delimiter, now we need to check that is followed strictly by a comma or brace
+                        while next_c and next_c.isspace():
+                            i += 1
+                            next_c = self.get_char_at(i)
+                        if next_c and next_c in [",", "}"]:
+                            rstring_delimiter_missing = False
+                    if rstring_delimiter_missing:
+                        self.log(
+                            "While parsing a string missing the left delimiter in object value context, we found a , or } and we couldn't determine that a right delimiter was present. Stopping here",
+                            "info",
+                        )
+                        break
             string_acc += char
             self.index += 1
             char = self.get_char_at()
@@ -386,6 +418,33 @@ class JSONParser:
                         "While parsing a string, we found a doubled quote, ignoring it",
                         "info",
                     )
+                elif missing_quotes and self.get_context() == "object_value":
+                    # In case of missing starting quote I need to check if the delimeter is the end or the beginning of a key
+                    i = 1
+                    next_c = self.get_char_at(i)
+                    while next_c and next_c not in [
+                        rstring_delimiter,
+                        lstring_delimiter,
+                    ]:
+                        i += 1
+                        next_c = self.get_char_at(i)
+                    if next_c:
+                        # We found a quote, now let's make sure there's a ":" following
+                        i += 1
+                        next_c = self.get_char_at(i)
+                        # found a delimiter, now we need to check that is followed strictly by a comma or brace
+                        while next_c and next_c.isspace():
+                            i += 1
+                            next_c = self.get_char_at(i)
+                        if next_c and next_c == ":":
+                            # Reset the cursor
+                            self.index -= 1
+                            char = self.get_char_at()
+                            self.log(
+                                "In a string with missing quotes and object value context, I found a delimeter but it turns out it was the beginning on the next key. Stopping here.",
+                                "info",
+                            )
+                            break
                 else:
                     # Check if eventually there is a rstring delimiter, otherwise we bail
                     i = 1
@@ -496,7 +555,8 @@ class JSONParser:
         number_str = ""
         number_chars = set("0123456789-.eE/,")
         char = self.get_char_at()
-        while char and char in number_chars:
+        is_array = self.get_context() == "array"
+        while char and char in number_chars and (char != "," or not is_array):
             number_str += char
             self.index += 1
             char = self.get_char_at()

{json_repair-0.23.1.dist-info → json_repair-0.25.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: json_repair
-Version: 0.23.1
+Version: 0.25.0
 Summary: A package to repair broken json strings
 Author-email: Stefano Baccianella <4247706+mangiucugna@users.noreply.github.com>
 License: MIT License
@@ -45,6 +45,11 @@ If you find this library useful, you can help me by donating toward my monthly b
 
 ---
 
+# Demo
+If you are unsure if this library will fix your specific problem, or simply want your json validated online, you can visit the demo site on GitHub pages: https://mangiucugna.github.io/json_repair/
+
+---
+
 # Motivation
 Some LLMs are a bit iffy when it comes to returning well formed JSON data, sometimes they skip a parentheses and sometimes they add some words in it, because that's what an LLM does.
 Luckily, the mistakes LLMs make are simple enough to be fixed without destroying the content.
@@ -160,6 +165,7 @@ You will need owner access to this repository
 # Repair JSON in other programming languages
 - Typescript: https://github.com/josdejong/jsonrepair
 - Go: https://github.com/RealAlexandreAI/json-repair
+- Ruby: https://github.com/sashazykov/json-repair-rb
 ---
 ## Star History
 

json_repair-0.25.0.dist-info/RECORD

@@ -0,0 +1,7 @@
+json_repair/__init__.py,sha256=IIzSm1DsCRrr8seF3UeMZXwxcq-tE3j-8d1WBxvEJvE,178
+json_repair/json_repair.py,sha256=RyxtqoMopJvEwVestHDP_D6FnVkSt3bmmr1liYOQwHE,30976
+json_repair-0.25.0.dist-info/LICENSE,sha256=wrjQo8MhNrNCicXtMe3MHmS-fx8AmQk1ue8AQwiiFV8,1076
+json_repair-0.25.0.dist-info/METADATA,sha256=K6rBtBRbVuRQICWknDvDm3OnNq_bGrwT7K6U99PPYmg,7596
+json_repair-0.25.0.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
+json_repair-0.25.0.dist-info/top_level.txt,sha256=7-VZwZN2CgB_n0NlSLk-rEUFh8ug21lESbsblOYuZqw,12
+json_repair-0.25.0.dist-info/RECORD,,

json_repair-0.23.1.dist-info/RECORD

@@ -1,7 +0,0 @@
-json_repair/__init__.py,sha256=IIzSm1DsCRrr8seF3UeMZXwxcq-tE3j-8d1WBxvEJvE,178
-json_repair/json_repair.py,sha256=3OUrtLLPc0fK0_U6H5-2wEZp1I6GqF5u9K5It6VOIzM,27629
-json_repair-0.23.1.dist-info/LICENSE,sha256=wrjQo8MhNrNCicXtMe3MHmS-fx8AmQk1ue8AQwiiFV8,1076
-json_repair-0.23.1.dist-info/METADATA,sha256=ujJO8L3zXSI8SRrZSF8qkcAaM8aj7XHWgx88x_OAnmA,7333
-json_repair-0.23.1.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-json_repair-0.23.1.dist-info/top_level.txt,sha256=7-VZwZN2CgB_n0NlSLk-rEUFh8ug21lESbsblOYuZqw,12
-json_repair-0.23.1.dist-info/RECORD,,