json-repair 0.28.3__tar.gz → 0.29.0__tar.gz

{json_repair-0.28.3/src/json_repair.egg-info → json_repair-0.29.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: json_repair
-Version: 0.28.3
+Version: 0.29.0
 Summary: A package to repair broken json strings
 Author-email: Stefano Baccianella <4247706+mangiucugna@users.noreply.github.com>
 License: MIT License
@@ -45,6 +45,21 @@ This simple package can be used to fix an invalid json string. To know all cases
 
 Inspired by https://github.com/josdejong/jsonrepair
 
+---
+# How to cite
+If you are using this library in your academic work (as I know many folks are) please find the BibTex here:
+
+    @software{Baccianella_JSON_Repair_-_2024,
+        author = {Baccianella, Stefano},
+        month = aug,
+        title = {{JSON Repair - A python module to repair invalid JSON, commonly used to parse the output of LLMs}},
+        url = {https://github.com/mangiucugna/json_repair},
+        version = {0.28.3},
+        year = {2024}
+    }
+
+Thank you for citing my work and please send me a link to the paper if you can!
+
 ---
 # Offer me a beer
 If you find this library useful, you can help me by donating toward my monthly beer budget here: https://github.com/sponsors/mangiucugna
@@ -82,6 +97,18 @@ or just
 
     decoded_object = json_repair.repair_json(json_string, return_objects=True)
 
+### Avoid this antipattern
+Some users of this library adopt the following pattern:
+
+    obj = {}
+    try:
+        obj = json.loads(string)
+    except json.JSONDecodeError as e:
+        obj = json_repair.loads(string)
+        ...
+
+This is wasteful because `json_repair` will already verify for you if the JSON is valid, if you still want to do that then add `skip_json_loads=True` to the call as explained the section below.
+
 ### Read json from a file or file descriptor
 
 JSON repair provides also a drop-in replacement for `json.load()`:
@@ -122,6 +149,32 @@ 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\""`
+
+### Use json_repair from CLI
+
+Install the library for command-line with:
+```
+pipx install json-repair
+```
+then run
+```
+$ json_repair -h
+
+usage: json_repair [-h] [-i] [--ensure_ascii] [--indent INDENT] filename
+
+Repair and parse JSON files.
+
+positional arguments:
+  filename         The JSON file to repair
+
+options:
+  -h, --help       show this help message and exit
+  -i, --inline     Replace the file inline instead of returning the output to stdout
+  --ensure_ascii   Pass the ensure_ascii parameter to json.dumps()
+  --indent INDENT  Number of spaces for indentation (Default 2)
+```
+to learn how to use it
+
 ## Adding to requirements
 **Please pin this library only on the major version!**
 

{json_repair-0.28.3 → json_repair-0.29.0}/README.md

@@ -7,6 +7,21 @@ This simple package can be used to fix an invalid json string. To know all cases
 
 Inspired by https://github.com/josdejong/jsonrepair
 
+---
+# How to cite
+If you are using this library in your academic work (as I know many folks are) please find the BibTex here:
+
+    @software{Baccianella_JSON_Repair_-_2024,
+        author = {Baccianella, Stefano},
+        month = aug,
+        title = {{JSON Repair - A python module to repair invalid JSON, commonly used to parse the output of LLMs}},
+        url = {https://github.com/mangiucugna/json_repair},
+        version = {0.28.3},
+        year = {2024}
+    }
+
+Thank you for citing my work and please send me a link to the paper if you can!
+
 ---
 # Offer me a beer
 If you find this library useful, you can help me by donating toward my monthly beer budget here: https://github.com/sponsors/mangiucugna
@@ -44,6 +59,18 @@ or just
 
     decoded_object = json_repair.repair_json(json_string, return_objects=True)
 
+### Avoid this antipattern
+Some users of this library adopt the following pattern:
+
+    obj = {}
+    try:
+        obj = json.loads(string)
+    except json.JSONDecodeError as e:
+        obj = json_repair.loads(string)
+        ...
+
+This is wasteful because `json_repair` will already verify for you if the JSON is valid, if you still want to do that then add `skip_json_loads=True` to the call as explained the section below.
+
 ### Read json from a file or file descriptor
 
 JSON repair provides also a drop-in replacement for `json.load()`:
@@ -84,6 +111,32 @@ 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\""`
+
+### Use json_repair from CLI
+
+Install the library for command-line with:
+```
+pipx install json-repair
+```
+then run
+```
+$ json_repair -h
+
+usage: json_repair [-h] [-i] [--ensure_ascii] [--indent INDENT] filename
+
+Repair and parse JSON files.
+
+positional arguments:
+  filename         The JSON file to repair
+
+options:
+  -h, --help       show this help message and exit
+  -i, --inline     Replace the file inline instead of returning the output to stdout
+  --ensure_ascii   Pass the ensure_ascii parameter to json.dumps()
+  --indent INDENT  Number of spaces for indentation (Default 2)
+```
+to learn how to use it
+
 ## Adding to requirements
 **Please pin this library only on the major version!**
 

{json_repair-0.28.3 → json_repair-0.29.0}/pyproject.toml

@@ -3,7 +3,7 @@ requires = ["setuptools>=61.0"]
 build-backend = "setuptools.build_meta"
 [project]
 name = "json_repair"
-version = "0.28.3"
+version = "0.29.0"
 license = {file = "LICENSE"}
 authors = [
   { name="Stefano Baccianella", email="4247706+mangiucugna@users.noreply.github.com" },
@@ -29,3 +29,5 @@ pythonpath = [
 "pkgname" = ["py.typed"]
 [tool.setuptools.packages.find]
 where = ["src"]
+[project.scripts]
+json_repair = "json_repair.__main__:cli"

json_repair-0.29.0/src/json_repair/__main__.py

@@ -0,0 +1,4 @@
+from .json_repair import cli
+
+if __name__ == "__main__":
+    cli()

{json_repair-0.28.3 → json_repair-0.29.0}/src/json_repair/json_repair.py

@@ -22,26 +22,62 @@ If something is wrong (a missing parantheses or quotes for example) it will use
 All supported use cases are in the unit tests
 """
 
+import argparse
 import os
+import sys
 import json
 from typing import Any, Dict, List, Optional, Union, TextIO, Tuple, Literal
 
 
 class StringFileWrapper:
     # This is a trick to simplify the code, transform the filedescriptor handling into a string handling
-    def __init__(self, fd: TextIO) -> None:
+    def __init__(self, fd: TextIO, CHUNK_LENGTH: int) -> None:
         self.fd = fd
         self.length: int = 0
+        # Buffers are 1MB strings that are read from the file
+        # and kept in memory to keep reads low
+        self.buffers: dict[int, str] = {}
+        # CHUNK_LENGTH is in bytes
+        if not CHUNK_LENGTH or CHUNK_LENGTH < 2:
+            CHUNK_LENGTH = 1_000_000
+        self.buffer_length = CHUNK_LENGTH
+
+    def get_buffer(self, index: int) -> str:
+        if self.buffers.get(index) is None:
+            self.fd.seek(index * self.buffer_length)
+            self.buffers[index] = self.fd.read(self.buffer_length)
+            # Save memory by keeping max 2MB buffer chunks and min 2 chunks
+            if len(self.buffers) > max(2, 2_000_000 / self.buffer_length):
+                oldest_key = next(iter(self.buffers))
+                if oldest_key != index:
+                    self.buffers.pop(oldest_key)
+        return self.buffers[index]
 
     def __getitem__(self, index: Union[int, slice]) -> str:
+        # The buffer is an array that is seek like a RAM:
+        # self.buffers[index]: the row in the array of length 1MB, index is `i` modulo CHUNK_LENGTH
+        # self.buffures[index][j]: the column of the row that is `i` remainder CHUNK_LENGTH
         if isinstance(index, slice):
-            self.fd.seek(index.start)
-            value = self.fd.read(index.stop - index.start)
-            self.fd.seek(index.start)
-            return value
+            buffer_index = index.start // self.buffer_length
+            buffer_end = index.stop // self.buffer_length
+            if buffer_index == buffer_end:
+                return self.get_buffer(buffer_index)[
+                    index.start % self.buffer_length : index.stop % self.buffer_length
+                ]
+            else:
+                start_slice = self.get_buffer(buffer_index)[
+                    index.start % self.buffer_length :
+                ]
+                end_slice = self.get_buffer(buffer_end)[
+                    : index.stop % self.buffer_length
+                ]
+                middle_slices = [
+                    self.get_buffer(i) for i in range(buffer_index + 1, buffer_end)
+                ]
+                return start_slice + "".join(middle_slices) + end_slice
         else:
-            self.fd.seek(index)
-            return self.fd.read(1)
+            buffer_index = index // self.buffer_length
+            return self.get_buffer(buffer_index)[index % self.buffer_length]
 
     def __len__(self) -> int:
         if self.length < 1:
@@ -69,13 +105,14 @@ class JSONParser:
         json_str: Union[str, StringFileWrapper],
         json_fd: Optional[TextIO],
         logging: Optional[bool],
+        json_fd_chunk_length: int = 0,
     ) -> None:
         # The string to parse
         self.json_str = json_str
         # Alternatively, the file description with a json file in it
         if json_fd:
             # This is a trick we do to treat the file wrapper as an array
-            self.json_str = StringFileWrapper(json_fd)
+            self.json_str = StringFileWrapper(json_fd, json_fd_chunk_length)
         # Index is our iterator that will keep track of which character we are looking at right now
         self.index: int = 0
         # This is used in the object member parsing to manage the special cases of missing quotes in key or value
@@ -639,6 +676,7 @@ def repair_json(
     logging: bool = False,
     json_fd: Optional[TextIO] = None,
     ensure_ascii: bool = True,
+    chunk_length: int = 0,
 ) -> Union[JSONReturnType, Tuple[JSONReturnType, List[Dict[str, str]]]]:
     """
     Given a json formatted string, it will try to decode it and, if it fails, it will try to fix it.
@@ -647,7 +685,7 @@ def repair_json(
     When `skip_json_loads=True` is passed, it will not call the built-in json.loads() function
     When `logging=True` is passed, it will return a tuple with the repaired json and a log of all repair actions
     """
-    parser = JSONParser(json_str, json_fd, logging)
+    parser = JSONParser(json_str, json_fd, logging, chunk_length)
     if skip_json_loads:
         parsed_json = parser.parse()
     else:
@@ -683,7 +721,10 @@ def loads(
 
 
 def load(
-    fd: TextIO, skip_json_loads: bool = False, logging: bool = False
+    fd: TextIO,
+    skip_json_loads: bool = False,
+    logging: bool = False,
+    chunk_length: int = 0,
 ) -> Union[JSONReturnType, Tuple[JSONReturnType, List[Dict[str, str]]]]:
     """
     This function works like `json.load()` except that it will fix your JSON in the process.
@@ -691,6 +732,7 @@ def load(
     """
     return repair_json(
         json_fd=fd,
+        chunk_length=chunk_length,
         return_objects=True,
         skip_json_loads=skip_json_loads,
         logging=logging,
@@ -701,12 +743,62 @@ def from_file(
     filename: str,
     skip_json_loads: bool = False,
     logging: bool = False,
+    chunk_length: int = 0,
 ) -> Union[JSONReturnType, Tuple[JSONReturnType, List[Dict[str, str]]]]:
     """
     This function is a wrapper around `load()` so you can pass the filename as string
     """
     fd = open(filename)
-    jsonobj = load(fd, skip_json_loads, logging)
+    jsonobj = load(
+        fd=fd,
+        skip_json_loads=skip_json_loads,
+        logging=logging,
+        chunk_length=chunk_length,
+    )
     fd.close()
 
     return jsonobj
+
+
+def cli():  # pragma: no cover
+    parser = argparse.ArgumentParser(description="Repair and parse JSON files.")
+    parser.add_argument("filename", help="The JSON file to repair")
+    parser.add_argument(
+        "-i",
+        "--inline",
+        action="store_true",
+        help="Replace the file inline instead of returning the output to stdout",
+    )
+    parser.add_argument(
+        "--ensure_ascii",
+        action="store_true",
+        help="Pass the ensure_ascii parameter to json.dumps()",
+    )
+    parser.add_argument(
+        "--indent",
+        type=int,
+        default=2,
+        help="Number of spaces for indentation (Default 2)",
+    )
+
+    args = parser.parse_args()
+
+    ensure_ascii = False
+    if args.ensure_ascii:
+        ensure_ascii = True
+    try:
+        result = from_file(args.filename)
+
+        if args.inline:
+            fd = open(args.filename, mode="w")
+            json.dump(result, fd, indent=args.indent, ensure_ascii=ensure_ascii)
+            fd.close()
+        else:
+            print(json.dumps(result, indent=args.indent, ensure_ascii=ensure_ascii))
+    except Exception as e:
+        print(f"Error: {str(e)}", file=sys.stderr)
+        sys.exit(1)
+
+
+if __name__ == "__main__":  # pragma: no cover
+    cli()

{json_repair-0.28.3 → json_repair-0.29.0/src/json_repair.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: json_repair
-Version: 0.28.3
+Version: 0.29.0
 Summary: A package to repair broken json strings
 Author-email: Stefano Baccianella <4247706+mangiucugna@users.noreply.github.com>
 License: MIT License
@@ -45,6 +45,21 @@ This simple package can be used to fix an invalid json string. To know all cases
 
 Inspired by https://github.com/josdejong/jsonrepair
 
+---
+# How to cite
+If you are using this library in your academic work (as I know many folks are) please find the BibTex here:
+
+    @software{Baccianella_JSON_Repair_-_2024,
+        author = {Baccianella, Stefano},
+        month = aug,
+        title = {{JSON Repair - A python module to repair invalid JSON, commonly used to parse the output of LLMs}},
+        url = {https://github.com/mangiucugna/json_repair},
+        version = {0.28.3},
+        year = {2024}
+    }
+
+Thank you for citing my work and please send me a link to the paper if you can!
+
 ---
 # Offer me a beer
 If you find this library useful, you can help me by donating toward my monthly beer budget here: https://github.com/sponsors/mangiucugna
@@ -82,6 +97,18 @@ or just
 
     decoded_object = json_repair.repair_json(json_string, return_objects=True)
 
+### Avoid this antipattern
+Some users of this library adopt the following pattern:
+
+    obj = {}
+    try:
+        obj = json.loads(string)
+    except json.JSONDecodeError as e:
+        obj = json_repair.loads(string)
+        ...
+
+This is wasteful because `json_repair` will already verify for you if the JSON is valid, if you still want to do that then add `skip_json_loads=True` to the call as explained the section below.
+
 ### Read json from a file or file descriptor
 
 JSON repair provides also a drop-in replacement for `json.load()`:
@@ -122,6 +149,32 @@ 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\""`
+
+### Use json_repair from CLI
+
+Install the library for command-line with:
+```
+pipx install json-repair
+```
+then run
+```
+$ json_repair -h
+
+usage: json_repair [-h] [-i] [--ensure_ascii] [--indent INDENT] filename
+
+Repair and parse JSON files.
+
+positional arguments:
+  filename         The JSON file to repair
+
+options:
+  -h, --help       show this help message and exit
+  -i, --inline     Replace the file inline instead of returning the output to stdout
+  --ensure_ascii   Pass the ensure_ascii parameter to json.dumps()
+  --indent INDENT  Number of spaces for indentation (Default 2)
+```
+to learn how to use it
+
 ## Adding to requirements
 **Please pin this library only on the major version!**
 

{json_repair-0.28.3 → json_repair-0.29.0}/src/json_repair.egg-info/SOURCES.txt

@@ -2,11 +2,13 @@ LICENSE
 README.md
 pyproject.toml
 src/json_repair/__init__.py
+src/json_repair/__main__.py
 src/json_repair/json_repair.py
 src/json_repair/py.typed
 src/json_repair.egg-info/PKG-INFO
 src/json_repair.egg-info/SOURCES.txt
 src/json_repair.egg-info/dependency_links.txt
+src/json_repair.egg-info/entry_points.txt
 src/json_repair.egg-info/top_level.txt
 tests/test_coverage.py
 tests/test_json_repair.py

json_repair-0.29.0/src/json_repair.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+json_repair = json_repair.__main__:cli

json_repair-0.29.0/tests/test_json_repair.py

@@ -0,0 +1,265 @@
+from src.json_repair.json_repair import from_file, repair_json, loads
+
+def test_basic_types_valid():
+    assert repair_json("True", return_objects=True) == ""
+    assert repair_json("False", return_objects=True) == ""
+    assert repair_json("Null", return_objects=True) == ""
+    assert repair_json("1", return_objects=True) == 1
+    assert repair_json("[]", return_objects=True) == []
+    assert repair_json("[1, 2, 3, 4]", return_objects=True) == [1, 2, 3, 4]
+    assert repair_json("{}", return_objects=True) == {}
+    assert repair_json('{ "key": "value", "key2": 1, "key3": True }', return_objects=True) == { "key": "value", "key2": 1, "key3": True }
+
+def test_basic_types_invalid():
+    assert repair_json("true", return_objects=True) == True
+    assert repair_json("false", return_objects=True) == False
+    assert repair_json("null", return_objects=True) == None
+    assert repair_json("1.2", return_objects=True) == 1.2
+    assert repair_json("[", return_objects=True) == []
+    assert repair_json("[1, 2, 3, 4", return_objects=True) == [1, 2, 3, 4]
+    assert repair_json("{", return_objects=True) == {}
+    assert repair_json('{ "key": value, "key2": 1 "key3": null }', return_objects=True) == { "key": "value", "key2": 1, "key3": None }
+
+def test_valid_json():
+    assert (
+        repair_json('{"name": "John", "age": 30, "city": "New York"}')
+        == '{"name": "John", "age": 30, "city": "New York"}'
+    )
+    assert (
+        repair_json('{"employees":["John", "Anna", "Peter"]} ')
+        == '{"employees": ["John", "Anna", "Peter"]}'
+    )
+    assert repair_json('{"key": "value:value"}') == '{"key": "value:value"}'
+    assert (
+        repair_json('{"text": "The quick brown fox,"}')
+        == '{"text": "The quick brown fox,"}'
+    )
+    assert (
+        repair_json('{"text": "The quick brown fox won\'t jump"}')
+        == '{"text": "The quick brown fox won\'t jump"}'
+    )
+    assert repair_json('{"key": ""') == '{"key": ""}'
+    assert (
+        repair_json('{"key1": {"key2": [1, 2, 3]}}') == '{"key1": {"key2": [1, 2, 3]}}'
+    )
+    assert (
+        repair_json('{"key": 12345678901234567890}') == '{"key": 12345678901234567890}'
+    )
+    assert repair_json('{"key": "value\u263A"}') == '{"key": "value\\u263a"}'
+    assert repair_json('{"key": "value\\nvalue"}') == '{"key": "value\\nvalue"}'
+
+def test_brackets_edge_cases():
+    assert repair_json("[{]") == "[{}]"
+    assert repair_json("   {  }   ") == "{}"
+    assert repair_json("[") == "[]"
+    assert repair_json("]") == '""'
+    assert repair_json("{") == "{}"
+    assert repair_json("}") == '""'
+    assert repair_json('{"') == '{}'
+    assert repair_json('["') == '[]'
+    assert repair_json('{foo: [}') == '{"foo": []}'
+
+def test_general_edge_cases():
+    assert repair_json("\"") == '""'
+    assert repair_json("\n") == '""'
+    assert repair_json(" ") == '""'
+    assert repair_json("[[1\n\n]") == "[[1]]"
+    assert repair_json("string") == '""'
+    assert repair_json("stringbeforeobject {}") == '{}'
+
+def test_mixed_data_types(): 
+    assert repair_json('  {"key": true, "key2": false, "key3": null}') == '{"key": true, "key2": false, "key3": null}'
+    assert repair_json('{"key": TRUE, "key2": FALSE, "key3": Null}   ') == '{"key": true, "key2": false, "key3": null}'
+
+def test_missing_and_mixed_quotes():
+    assert repair_json("{'key': 'string', 'key2': false, \"key3\": null, \"key4\": unquoted}") == '{"key": "string", "key2": false, "key3": null, "key4": "unquoted"}'
+    assert (
+        repair_json('{"name": "John", "age": 30, "city": "New York')
+        == '{"name": "John", "age": 30, "city": "New York"}'
+    )
+    assert (
+        repair_json('{"name": "John", "age": 30, city: "New York"}')
+        == '{"name": "John", "age": 30, "city": "New York"}'
+    )
+    assert (
+        repair_json('{"name": "John", "age": 30, "city": New York}')
+        == '{"name": "John", "age": 30, "city": "New York"}'
+    )
+    assert (
+        repair_json('{"name": John, "age": 30, "city": "New York"}')
+        == '{"name": "John", "age": 30, "city": "New York"}'
+    )
+    assert repair_json('{“slanted_delimiter”: "value"}') == '{"slanted_delimiter": "value"}'
+    assert (
+        repair_json('{"name": "John", "age": 30, "city": "New')
+        == '{"name": "John", "age": 30, "city": "New"}'
+    )
+    assert repair_json('[{"key": "value", COMMENT "notes": "lorem "ipsum", sic." }]') == '[{"key": "value", "notes": "lorem \\"ipsum\\", sic."}]'
+    assert repair_json('{"key": ""value"}') == '{"key": "value"}'
+    assert repair_json('{"key": "value", 5: "value"}') == '{"key": "value", "5": "value"}'
+    assert repair_json('{"foo": "\\"bar\\""') == '{"foo": "\\"bar\\""}'
+    assert repair_json('{"" key":"val"') == '{" key": "val"}'
+    assert repair_json('{"key": value "key2" : "value2" ') == '{"key": "value", "key2": "value2"}'
+
+def test_array_edge_cases():
+    assert repair_json("[1, 2, 3,") == "[1, 2, 3]"
+    assert repair_json("[1, 2, 3, ...]") == "[1, 2, 3]"
+    assert repair_json("[1, 2, ... , 3]") == "[1, 2, 3]"
+    assert repair_json("[1, 2, '...', 3]") == '[1, 2, "...", 3]'
+    assert repair_json("[true, false, null, ...]") == '[true, false, null]'
+    assert repair_json('["a" "b" "c" 1') == '["a", "b", "c", 1]'
+    assert repair_json('{"employees":["John", "Anna",') == '{"employees": ["John", "Anna"]}'
+    assert repair_json('{"employees":["John", "Anna", "Peter') == '{"employees": ["John", "Anna", "Peter"]}'
+    assert repair_json('{"key1": {"key2": [1, 2, 3') == '{"key1": {"key2": [1, 2, 3]}}'
+    
+def test_escaping():
+    assert repair_json("'\"'") == '""'
+    assert repair_json("{\"key\": 'string\"\n\t\le'") == '{"key": "string\\"\\n\\tle"}'
+    assert repair_json(r'{"real_content": "Some string: Some other string \t Some string <a href=\"https://domain.com\">Some link</a>"') == r'{"real_content": "Some string: Some other string \t Some string <a href=\"https://domain.com\">Some link</a>"}'
+    assert repair_json('{"key_1\n": "value"}') == '{"key_1": "value"}'
+    assert repair_json('{"key\t_": "value"}') == '{"key\\t_": "value"}'
+    
+    
+def test_object_edge_cases():
+    assert repair_json('{       ') == '{}'
+    assert repair_json('{"": "value"') == '{"": "value"}'
+    assert repair_json('{"value_1": true, COMMENT "value_2": "data"}') == '{"value_1": true, "value_2": "data"}'
+    assert repair_json('{"value_1": true, SHOULD_NOT_EXIST "value_2": "data" AAAA }') == '{"value_1": true, "value_2": "data"}'
+    assert repair_json('{"" : true, "key2": "value2"}') == '{"": true, "key2": "value2"}'
+    assert repair_json('{""answer"":[{""traits"":''Female aged 60+'',""answer1"":""5""}]}') == '{"answer": [{"traits": "Female aged 60+", "answer1": "5"}]}'
+    assert repair_json('{ "words": abcdef", "numbers": 12345", "words2": ghijkl" }') == '{"words": "abcdef", "numbers": 12345, "words2": "ghijkl"}'
+    assert repair_json('''{"number": 1,"reason": "According...""ans": "YES"}''') == '{"number": 1, "reason": "According...", "ans": "YES"}'
+    assert repair_json('''{ "a" : "{ b": {} }" }''') == '{"a": "{ b"}'
+    assert repair_json("""{"b": "xxxxx" true}""") == '{"b": "xxxxx"}'
+    assert repair_json('{"key": "Lorem "ipsum" s,"}') == '{"key": "Lorem \\"ipsum\\" s,"}'
+    assert repair_json('{"lorem": ipsum, sic, datum.",}') == '{"lorem": "ipsum, sic, datum."}'
+    assert repair_json('{"lorem": sic tamet. "ipsum": sic tamet, quick brown fox. "sic": ipsum}') == '{"lorem": "sic tamet.", "ipsum": "sic tamet", "sic": "ipsum"}'
+    assert repair_json('{"key":value, " key2":"value2" }') == '{"key": "value", " key2": "value2"}'
+    assert repair_json('{"key":value "key2":"value2" }') == '{"key": "value", "key2": "value2"}'
+
+def test_number_edge_cases():
+    assert repair_json(' - { "test_key": ["test_value", "test_value2"] }') == '{"test_key": ["test_value", "test_value2"]}'
+    assert repair_json('{"key": 1/3}') == '{"key": "1/3"}'
+    assert repair_json('{"key": .25}') == '{"key": 0.25}'
+    assert repair_json('{"here": "now", "key": 1/3, "foo": "bar"}') == '{"here": "now", "key": "1/3", "foo": "bar"}'
+    assert repair_json('{"key": 12345/67890}') == '{"key": "12345/67890"}'
+    assert repair_json('[105,12') == '[105, 12]'
+    assert repair_json('{"key", 105,12,') == '{"key": "105,12"}'
+    assert repair_json('{"key": 1/3, "foo": "bar"}') == '{"key": "1/3", "foo": "bar"}'
+    assert repair_json('{"key": 10-20}') == '{"key": "10-20"}'
+    assert repair_json('{"key": 1.1.1}') == '{"key": "1.1.1"}'
+    assert repair_json('[- ') == '[]'
+
+def test_markdown():
+    assert repair_json('{ "content": "[LINK]("https://google.com")" }') == '{"content": "[LINK](\\"https://google.com\\")"}'
+    assert repair_json('{ "content": "[LINK](" }') == '{"content": "[LINK]("}'
+    assert repair_json('{ "content": "[LINK](", "key": true }') == '{"content": "[LINK](", "key": true}'
+
+def test_leading_trailing_characters():
+    assert repair_json('````{ "key": "value" }```') == '{"key": "value"}'
+    assert repair_json("""{    "a": "",    "b": [ { "c": 1} ] \n}```""") == '{"a": "", "b": [{"c": 1}]}'
+    assert repair_json("Based on the information extracted, here is the filled JSON output: ```json { 'a': 'b' } ```") == '{"a": "b"}'
+    assert repair_json("""
+                       The next 64 elements are:
+                       ```json
+                       { "key": "value" }
+                       ```""") == '{"key": "value"}'
+def test_multiple_jsons():
+    assert repair_json("[]{}") == "[[], {}]"
+    assert repair_json("{}[]{}") == "[{}, [], {}]"
+    assert repair_json('{"key":"value"}[1,2,3,True]') == '[{"key": "value"}, [1, 2, 3, true]]'
+    assert repair_json('lorem ```json {"key":"value"} ``` ipsum ```json [1,2,3,True] ``` 42') == '[{"key": "value"}, [1, 2, 3, true]]'
+
+def test_repair_json_with_objects():
+    # Test with valid JSON strings
+    assert repair_json("[]", return_objects=True) == []
+    assert repair_json("{}", return_objects=True) == {}
+    assert repair_json('{"key": true, "key2": false, "key3": null}', return_objects=True) == {"key": True, "key2": False, "key3": None}
+    assert repair_json('{"name": "John", "age": 30, "city": "New York"}', return_objects=True) == {
+        "name": "John",
+        "age": 30,
+        "city": "New York",
+    }
+    assert repair_json("[1, 2, 3, 4]", return_objects=True) == [1, 2, 3, 4]
+    assert repair_json('{"employees":["John", "Anna", "Peter"]} ', return_objects=True) == {
+        "employees": ["John", "Anna", "Peter"]
+    }
+    assert repair_json('''
+{
+  "resourceType": "Bundle",
+  "id": "1",
+  "type": "collection",
+  "entry": [
+    {
+      "resource": {
+        "resourceType": "Patient",
+        "id": "1",
+        "name": [
+          {"use": "official", "family": "Corwin", "given": ["Keisha", "Sunny"], "prefix": ["Mrs."},
+          {"use": "maiden", "family": "Goodwin", "given": ["Keisha", "Sunny"], "prefix": ["Mrs."]}
+        ]
+      }
+    }
+  ]
+}
+''', return_objects=True) == {"resourceType": "Bundle", "id": "1", "type": "collection", "entry": [{"resource": {"resourceType": "Patient", "id": "1", "name": [{"use": "official", "family": "Corwin", "given": ["Keisha", "Sunny"], "prefix": ["Mrs."]}, {"use": "maiden", "family": "Goodwin", "given": ["Keisha", "Sunny"], "prefix": ["Mrs."]}]}}]}
+    assert repair_json('{\n"html": "<h3 id="aaa">Waarom meer dan 200 Technical Experts - "Passie voor techniek"?</h3>"}', return_objects=True) == {'html': '<h3 id="aaa">Waarom meer dan 200 Technical Experts - "Passie voor techniek"?</h3>'}
+    assert repair_json("""
+        [
+            {
+                "foo": "Foo bar baz",
+                "tag": "#foo-bar-baz"
+            },
+            {
+                "foo": "foo bar "foobar" foo bar baz.",
+                "tag": "#foo-bar-foobar"
+            }
+        ]
+        """, return_objects=True) == [{"foo": "Foo bar baz", "tag": "#foo-bar-baz"},{"foo": "foo bar \"foobar\" foo bar baz.", "tag": "#foo-bar-foobar" }]
+
+def test_repair_json_skip_json_loads():
+    assert repair_json('{"key": true, "key2": false, "key3": null}', skip_json_loads=True) == '{"key": true, "key2": false, "key3": null}'
+    assert repair_json('{"key": true, "key2": false, "key3": null}', return_objects=True, skip_json_loads=True) == {"key": True, "key2": False, "key3": None}
+    assert repair_json('{"key": true, "key2": false, "key3": }', skip_json_loads=True) == '{"key": true, "key2": false, "key3": ""}'
+    assert loads('{"key": true, "key2": false, "key3": }', skip_json_loads=True) == {"key": True, "key2": False, "key3": ""}
+
+
+def test_repair_json_from_file():
+    import os.path
+    import pathlib
+    import tempfile
+
+    path = pathlib.Path(__file__).parent.resolve()
+
+    # Use chunk_length 2 to test the buffering feature
+    assert from_file(filename=os.path.join(path,"invalid.json")) == [{"_id": "655b66256574f09bdae8abe8", "index": 0, "guid": "31082ae3-b0f3-4406-90f4-cc450bd4379d", "isActive": False, "balance": "$2,562.78", "picture": "http://placehold.it/32x32", "age": 32, "eyeColor": "brown", "name": "Glover Rivas", "gender": "male", "company": "EMPIRICA", "email": "gloverrivas@empirica.com", "phone": "+1 (842) 507-3063", "address": "536 Montague Terrace, Jenkinsville, Kentucky, 2235", "about": "Mollit consectetur excepteur voluptate tempor dolore ullamco enim irure ullamco non enim officia. Voluptate occaecat proident laboris ea Lorem cupidatat reprehenderit nisi nisi aliqua. Amet nulla ipsum deserunt excepteur amet ad aute aute ex. Et enim minim sit veniam est quis dolor nisi sunt quis eiusmod in. Amet eiusmod cillum sunt occaecat dolor laboris voluptate in eiusmod irure aliqua duis.", "registered": "2023-11-18T09:32:36 -01:00", "latitude": 36.26102, "longitude": -91.304608, "tags": ["non", "tempor", "do", "ullamco", "dolore", "sunt", "ipsum"], "friends": [{"id": 0, "name": "Cara Shepherd"}, {"id": 1, "name": "Mason Farley"}, {"id": 2, "name": "Harriet Cochran"}], "greeting": "Hello, Glover Rivas! You have 7 unread messages.", "favoriteFruit": "strawberry"}, {"_id": "655b662585364bc57278bb6f", "index": 1, "guid": "0dea7a3a-f812-4dde-b78d-7a9b58e5da05", "isActive": True, "balance": "$1,359.48", "picture": "http://placehold.it/32x32", "age": 38, "eyeColor": "brown", "name": "Brandi Moreno", "gender": "female", "company": "MARQET", "email": "brandimoreno@marqet.com", "phone": "+1 (850) 434-2077", "address": "537 Doone Court, Waiohinu, Michigan, 3215", "about": "Irure proident adipisicing do Lorem do incididunt in laborum in eiusmod eiusmod ad elit proident. Eiusmod dolor ex magna magna occaecat. Nulla deserunt velit ex exercitation et irure sunt. Cupidatat ut excepteur ea quis labore sint cupidatat incididunt amet eu consectetur cillum ipsum proident. Occaecat exercitation aute laborum dolor proident reprehenderit laborum in voluptate culpa. Exercitation nulla adipisicing culpa aute est deserunt ea nisi deserunt consequat occaecat ut et non. Incididunt ex exercitation dolor dolor anim cillum dolore.", "registered": "2015-09-03T11:47:15 -02:00", "latitude": -19.768953, "longitude": 8.948458, "tags": ["laboris", "occaecat", "laborum", "laborum", "ex", "cillum", "occaecat"], "friends": [{"id": 0, "name": "Erna Kelly"}, {"id": 1, "name": "Black Mays"}, {"id": 2, "name": "Davis Buck"}], "greeting": "Hello, Brandi Moreno! You have 1 unread messages.", "favoriteFruit": "apple"}, {"_id": "655b6625870da431bcf5e0c2", "index": 2, "guid": "b17f6e3f-c898-4334-abbf-05cf222f143b", "isActive": False, "balance": "$1,493.77", "picture": "http://placehold.it/32x32", "age": 20, "eyeColor": "brown", "name": "Moody Meadows", "gender": "male", "company": "OPTIQUE", "email": "moodymeadows@optique.com", "phone": "+1 (993) 566-3041", "address": "766 Osborn Street, Bath, Maine, 7666", "about": "Non commodo excepteur nostrud qui adipisicing aliquip dolor minim nulla culpa proident. In ad cupidatat ea mollit ex est do deserunt proident nostrud. Cillum id id eiusmod amet exercitation nostrud cillum sunt deserunt dolore deserunt eiusmod mollit. Ut ex tempor ad laboris voluptate labore id officia fugiat exercitation amet.", "registered": "2015-01-16T02:48:28 -01:00", "latitude": -25.847327, "longitude": 63.95991, "tags": ["aute", "commodo", "adipisicing", "nostrud", "duis", "mollit", "ut"], "friends": [{"id": 0, "name": "Lacey Cash"}, {"id": 1, "name": "Gabrielle Harmon"}, {"id": 2, "name": "Ellis Lambert"}], "greeting": "Hello, Moody Meadows! You have 4 unread messages.", "favoriteFruit": "strawberry"}, {"_id": "655b6625f3e1bf422220854e", "index": 3, "guid": "92229883-2bfd-4974-a08c-1b506b372e46", "isActive": False, "balance": "$2,215.34", "picture": "http://placehold.it/32x32", "age": 22, "eyeColor": "brown", "name": "Heath Nguyen", "gender": "male", "company": "BLEENDOT", "email": "heathnguyen@bleendot.com", "phone": "+1 (989) 512-2797", "address": "135 Milton Street, Graniteville, Nebraska, 276", "about": "Consequat aliquip irure Lorem cupidatat nulla magna ullamco nulla voluptate adipisicing anim consectetur tempor aliquip. Magna aliqua nulla eu tempor esse proident. Proident fugiat ad ex Lorem reprehenderit dolor aliquip labore labore aliquip. Deserunt aute enim ea minim officia anim culpa sint commodo. Cillum consectetur excepteur aliqua exercitation Lorem veniam voluptate.", "registered": "2016-07-06T01:31:07 -02:00", "latitude": -60.997048, "longitude": -102.397885, "tags": ["do", "ad", "consequat", "irure", "tempor", "elit", "minim"], "friends": [{"id": 0, "name": "Walker Hernandez"}, {"id": 1, "name": "Maria Lane"}, {"id": 2, "name": "Mcknight Barron"}], "greeting": "Hello, Heath Nguyen! You have 4 unread messages.", "favoriteFruit": "apple"}, {"_id": "655b6625519a5b5e4b6742bf", "index": 4, "guid": "c5dc685f-6d0d-4173-b4cf-f5df29a1e8ef", "isActive": True, "balance": "$1,358.90", "picture": "http://placehold.it/32x32", "age": 33, "eyeColor": "brown", "name": "Deidre Duke", "gender": "female", "company": "OATFARM", "email": "deidreduke@oatfarm.com", "phone": "+1 (875) 587-3256", "address": "487 Schaefer Street, Wattsville, West Virginia, 4506", "about": "Laboris eu nulla esse magna sit eu deserunt non est aliqua exercitation commodo. Ad occaecat qui qui laborum dolore anim Lorem. Est qui occaecat irure enim deserunt enim aliqua ex deserunt incididunt esse. Quis in minim laboris proident non mollit. Magna ea do labore commodo. Et elit esse esse occaecat officia ipsum nisi.", "registered": "2021-09-12T04:17:08 -02:00", "latitude": 68.609781, "longitude": -87.509134, "tags": ["mollit", "cupidatat", "irure", "sit", "consequat", "anim", "fugiat"], "friends": [{"id": 0, "name": "Bean Paul"}, {"id": 1, "name": "Cochran Hubbard"}, {"id": 2, "name": "Rodgers Atkinson"}], "greeting": "Hello, Deidre Duke! You have 6 unread messages.", "favoriteFruit": "apple"}, {"_id": "655b6625a19b3f7e5f82f0ea", "index": 5, "guid": "75f3c264-baa1-47a0-b21c-4edac23d9935", "isActive": True, "balance": "$3,554.36", "picture": "http://placehold.it/32x32", "age": 26, "eyeColor": "blue", "name": "Lydia Holland", "gender": "female", "company": "ESCENTA", "email": "lydiaholland@escenta.com", "phone": "+1 (927) 482-3436", "address": "554 Rockaway Parkway, Kohatk, Montana, 6316", "about": "Consectetur ea est labore commodo laborum mollit pariatur non enim. Est dolore et non laboris tempor. Ea incididunt ut adipisicing cillum labore officia tempor eiusmod commodo. Cillum fugiat ex consectetur ut nostrud anim nostrud exercitation ut duis in ea. Eu et id fugiat est duis eiusmod ullamco quis officia minim sint ea nisi in.", "registered": "2018-03-13T01:48:56 -01:00", "latitude": -88.495799, "longitude": 71.840667, "tags": ["veniam", "minim", "consequat", "consequat", "incididunt", "consequat", "elit"], "friends": [{"id": 0, "name": "Debra Massey"}, {"id": 1, "name": "Weiss Savage"}, {"id": 2, "name": "Shannon Guerra"}], "greeting": "Hello, Lydia Holland! You have 5 unread messages.", "favoriteFruit": "banana"}]
+    assert from_file(filename=os.path.join(path,"invalid.json"), chunk_length=2) == [{"_id": "655b66256574f09bdae8abe8", "index": 0, "guid": "31082ae3-b0f3-4406-90f4-cc450bd4379d", "isActive": False, "balance": "$2,562.78", "picture": "http://placehold.it/32x32", "age": 32, "eyeColor": "brown", "name": "Glover Rivas", "gender": "male", "company": "EMPIRICA", "email": "gloverrivas@empirica.com", "phone": "+1 (842) 507-3063", "address": "536 Montague Terrace, Jenkinsville, Kentucky, 2235", "about": "Mollit consectetur excepteur voluptate tempor dolore ullamco enim irure ullamco non enim officia. Voluptate occaecat proident laboris ea Lorem cupidatat reprehenderit nisi nisi aliqua. Amet nulla ipsum deserunt excepteur amet ad aute aute ex. Et enim minim sit veniam est quis dolor nisi sunt quis eiusmod in. Amet eiusmod cillum sunt occaecat dolor laboris voluptate in eiusmod irure aliqua duis.", "registered": "2023-11-18T09:32:36 -01:00", "latitude": 36.26102, "longitude": -91.304608, "tags": ["non", "tempor", "do", "ullamco", "dolore", "sunt", "ipsum"], "friends": [{"id": 0, "name": "Cara Shepherd"}, {"id": 1, "name": "Mason Farley"}, {"id": 2, "name": "Harriet Cochran"}], "greeting": "Hello, Glover Rivas! You have 7 unread messages.", "favoriteFruit": "strawberry"}, {"_id": "655b662585364bc57278bb6f", "index": 1, "guid": "0dea7a3a-f812-4dde-b78d-7a9b58e5da05", "isActive": True, "balance": "$1,359.48", "picture": "http://placehold.it/32x32", "age": 38, "eyeColor": "brown", "name": "Brandi Moreno", "gender": "female", "company": "MARQET", "email": "brandimoreno@marqet.com", "phone": "+1 (850) 434-2077", "address": "537 Doone Court, Waiohinu, Michigan, 3215", "about": "Irure proident adipisicing do Lorem do incididunt in laborum in eiusmod eiusmod ad elit proident. Eiusmod dolor ex magna magna occaecat. Nulla deserunt velit ex exercitation et irure sunt. Cupidatat ut excepteur ea quis labore sint cupidatat incididunt amet eu consectetur cillum ipsum proident. Occaecat exercitation aute laborum dolor proident reprehenderit laborum in voluptate culpa. Exercitation nulla adipisicing culpa aute est deserunt ea nisi deserunt consequat occaecat ut et non. Incididunt ex exercitation dolor dolor anim cillum dolore.", "registered": "2015-09-03T11:47:15 -02:00", "latitude": -19.768953, "longitude": 8.948458, "tags": ["laboris", "occaecat", "laborum", "laborum", "ex", "cillum", "occaecat"], "friends": [{"id": 0, "name": "Erna Kelly"}, {"id": 1, "name": "Black Mays"}, {"id": 2, "name": "Davis Buck"}], "greeting": "Hello, Brandi Moreno! You have 1 unread messages.", "favoriteFruit": "apple"}, {"_id": "655b6625870da431bcf5e0c2", "index": 2, "guid": "b17f6e3f-c898-4334-abbf-05cf222f143b", "isActive": False, "balance": "$1,493.77", "picture": "http://placehold.it/32x32", "age": 20, "eyeColor": "brown", "name": "Moody Meadows", "gender": "male", "company": "OPTIQUE", "email": "moodymeadows@optique.com", "phone": "+1 (993) 566-3041", "address": "766 Osborn Street, Bath, Maine, 7666", "about": "Non commodo excepteur nostrud qui adipisicing aliquip dolor minim nulla culpa proident. In ad cupidatat ea mollit ex est do deserunt proident nostrud. Cillum id id eiusmod amet exercitation nostrud cillum sunt deserunt dolore deserunt eiusmod mollit. Ut ex tempor ad laboris voluptate labore id officia fugiat exercitation amet.", "registered": "2015-01-16T02:48:28 -01:00", "latitude": -25.847327, "longitude": 63.95991, "tags": ["aute", "commodo", "adipisicing", "nostrud", "duis", "mollit", "ut"], "friends": [{"id": 0, "name": "Lacey Cash"}, {"id": 1, "name": "Gabrielle Harmon"}, {"id": 2, "name": "Ellis Lambert"}], "greeting": "Hello, Moody Meadows! You have 4 unread messages.", "favoriteFruit": "strawberry"}, {"_id": "655b6625f3e1bf422220854e", "index": 3, "guid": "92229883-2bfd-4974-a08c-1b506b372e46", "isActive": False, "balance": "$2,215.34", "picture": "http://placehold.it/32x32", "age": 22, "eyeColor": "brown", "name": "Heath Nguyen", "gender": "male", "company": "BLEENDOT", "email": "heathnguyen@bleendot.com", "phone": "+1 (989) 512-2797", "address": "135 Milton Street, Graniteville, Nebraska, 276", "about": "Consequat aliquip irure Lorem cupidatat nulla magna ullamco nulla voluptate adipisicing anim consectetur tempor aliquip. Magna aliqua nulla eu tempor esse proident. Proident fugiat ad ex Lorem reprehenderit dolor aliquip labore labore aliquip. Deserunt aute enim ea minim officia anim culpa sint commodo. Cillum consectetur excepteur aliqua exercitation Lorem veniam voluptate.", "registered": "2016-07-06T01:31:07 -02:00", "latitude": -60.997048, "longitude": -102.397885, "tags": ["do", "ad", "consequat", "irure", "tempor", "elit", "minim"], "friends": [{"id": 0, "name": "Walker Hernandez"}, {"id": 1, "name": "Maria Lane"}, {"id": 2, "name": "Mcknight Barron"}], "greeting": "Hello, Heath Nguyen! You have 4 unread messages.", "favoriteFruit": "apple"}, {"_id": "655b6625519a5b5e4b6742bf", "index": 4, "guid": "c5dc685f-6d0d-4173-b4cf-f5df29a1e8ef", "isActive": True, "balance": "$1,358.90", "picture": "http://placehold.it/32x32", "age": 33, "eyeColor": "brown", "name": "Deidre Duke", "gender": "female", "company": "OATFARM", "email": "deidreduke@oatfarm.com", "phone": "+1 (875) 587-3256", "address": "487 Schaefer Street, Wattsville, West Virginia, 4506", "about": "Laboris eu nulla esse magna sit eu deserunt non est aliqua exercitation commodo. Ad occaecat qui qui laborum dolore anim Lorem. Est qui occaecat irure enim deserunt enim aliqua ex deserunt incididunt esse. Quis in minim laboris proident non mollit. Magna ea do labore commodo. Et elit esse esse occaecat officia ipsum nisi.", "registered": "2021-09-12T04:17:08 -02:00", "latitude": 68.609781, "longitude": -87.509134, "tags": ["mollit", "cupidatat", "irure", "sit", "consequat", "anim", "fugiat"], "friends": [{"id": 0, "name": "Bean Paul"}, {"id": 1, "name": "Cochran Hubbard"}, {"id": 2, "name": "Rodgers Atkinson"}], "greeting": "Hello, Deidre Duke! You have 6 unread messages.", "favoriteFruit": "apple"}, {"_id": "655b6625a19b3f7e5f82f0ea", "index": 5, "guid": "75f3c264-baa1-47a0-b21c-4edac23d9935", "isActive": True, "balance": "$3,554.36", "picture": "http://placehold.it/32x32", "age": 26, "eyeColor": "blue", "name": "Lydia Holland", "gender": "female", "company": "ESCENTA", "email": "lydiaholland@escenta.com", "phone": "+1 (927) 482-3436", "address": "554 Rockaway Parkway, Kohatk, Montana, 6316", "about": "Consectetur ea est labore commodo laborum mollit pariatur non enim. Est dolore et non laboris tempor. Ea incididunt ut adipisicing cillum labore officia tempor eiusmod commodo. Cillum fugiat ex consectetur ut nostrud anim nostrud exercitation ut duis in ea. Eu et id fugiat est duis eiusmod ullamco quis officia minim sint ea nisi in.", "registered": "2018-03-13T01:48:56 -01:00", "latitude": -88.495799, "longitude": 71.840667, "tags": ["veniam", "minim", "consequat", "consequat", "incididunt", "consequat", "elit"], "friends": [{"id": 0, "name": "Debra Massey"}, {"id": 1, "name": "Weiss Savage"}, {"id": 2, "name": "Shannon Guerra"}], "greeting": "Hello, Lydia Holland! You have 5 unread messages.", "favoriteFruit": "banana"}]
+
+    
+    # Create a temporary file
+    temp_fd, temp_path = tempfile.mkstemp(suffix=".json")
+    try:
+        # Write content to the temporary file
+        with os.fdopen(temp_fd, 'w') as tmp:
+            tmp.write("{key:value}")
+        assert from_file(filename=temp_path, logging=True) == ({'key': 'value'}, [{'text': 'While parsing a string, we found a literal instead of a quote', 'context': '{key:value}'}, {'text': 'While parsing a string, we found no starting quote. Will add the quote back', 'context': '{key:value}'}, {'context': '{key:value}', 'text': 'While parsing a string missing the left delimiter in object key context, we found a :, stopping here',}, {'text': 'While parsing a string, we missed the closing quote, ignoring', 'context': '{key:value}'}, {'text': 'While parsing a string, we found a literal instead of a quote', 'context': '{key:value}'}, {'text': 'While parsing a string, we found no starting quote. Will add the quote back', 'context': '{key:value}'}, {'context': '{key:value}', 'text': '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'}, {'text': 'While parsing a string, we missed the closing quote, ignoring', 'context': '{key:value}'}])
+        assert from_file(filename=temp_path, logging=True, chunk_length=2) == ({'key': 'value'}, [{'text': 'While parsing a string, we found a literal instead of a quote', 'context': '{key:value}'}, {'text': 'While parsing a string, we found no starting quote. Will add the quote back', 'context': '{key:value}'}, {'context': '{key:value}', 'text': 'While parsing a string missing the left delimiter in object key context, we found a :, stopping here',}, {'text': 'While parsing a string, we missed the closing quote, ignoring', 'context': '{key:value}'}, {'text': 'While parsing a string, we found a literal instead of a quote', 'context': '{key:value}'}, {'text': 'While parsing a string, we found no starting quote. Will add the quote back', 'context': '{key:value}'}, {'context': '{key:value}', 'text': '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'}, {'text': 'While parsing a string, we missed the closing quote, ignoring', 'context': '{key:value}'}])
+    finally:
+        # Clean up - delete the temporary file
+        os.remove(temp_path)
+
+    # Create a temporary file
+    temp_fd, temp_path = tempfile.mkstemp(suffix=".json")
+    try:
+        # Write content to the temporary file
+        with os.fdopen(temp_fd, 'w') as tmp:
+            tmp.write('x' * 5 * 1024 * 1024) # 5 MB
+        assert from_file(filename=temp_path, logging=True) == ('', [])
+        
+    finally:
+        # Clean up - delete the temporary file
+        os.remove(temp_path)
+
+
+def test_ensure_ascii():
+    assert repair_json("{'test_中国人_ascii':'统一码'}", ensure_ascii=False) == '{"test_中国人_ascii": "统一码"}'

{json_repair-0.28.3 → json_repair-0.29.0}/tests/test_performance.py

@@ -19,7 +19,7 @@ def test_true_true_correct(benchmark):
   mean_time = benchmark.stats.get("median")
 
   # Define your time threshold in seconds
-  max_time = 14 / 10 ** 4  # 1.4 millisecond
+  max_time = 15 / 10 ** 4  # 1.5 millisecond
 
   # 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"
@@ -31,7 +31,7 @@ def test_true_true_incorrect(benchmark):
   mean_time = benchmark.stats.get("median")
 
   # Define your time threshold in seconds
-  max_time = 14 / 10 ** 4  # 1.4 millisecond
+  max_time = 15 / 10 ** 4  # 1.5 millisecond
 
   # 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"
@@ -53,7 +53,7 @@ def test_true_false_incorrect(benchmark):
   mean_time = benchmark.stats.get("median")
 
   # Define your time threshold in seconds
-  max_time = 14 / 10 ** 4  # 1.4 millisecond
+  max_time = 15 / 10 ** 4  # 1.5 millisecond
 
   # 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"
@@ -64,7 +64,7 @@ def test_false_true_correct(benchmark):
   mean_time = benchmark.stats.get("median")
 
   # Define your time threshold in seconds
-  max_time = 14 / 10 ** 4  # 1.4 millisecond
+  max_time = 15 / 10 ** 4  # 1.5 millisecond
 
   # 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"
@@ -75,7 +75,7 @@ def test_false_true_incorrect(benchmark):
   mean_time = benchmark.stats.get("median")
   
   # Define your time threshold in seconds
-  max_time = 14 / 10 ** 4  # 1.4 millisecond
+  max_time = 15 / 10 ** 4  # 1.5 millisecond
 
   # 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"

json_repair-0.28.3/tests/test_json_repair.py

@@ -1,247 +0,0 @@
-from src.json_repair.json_repair import from_file, repair_json, loads
-
-def test_basic_types_valid():
-    assert repair_json("True", return_objects=True) == ""
-    assert repair_json("False", return_objects=True) == ""
-    assert repair_json("Null", return_objects=True) == ""
-    assert repair_json("1", return_objects=True) == 1
-    assert repair_json("[]", return_objects=True) == []
-    assert repair_json("[1, 2, 3, 4]", return_objects=True) == [1, 2, 3, 4]
-    assert repair_json("{}", return_objects=True) == {}
-    assert repair_json('{ "key": "value", "key2": 1, "key3": True }', return_objects=True) == { "key": "value", "key2": 1, "key3": True }
-
-def test_basic_types_invalid():
-    assert repair_json("true", return_objects=True) == True
-    assert repair_json("false", return_objects=True) == False
-    assert repair_json("null", return_objects=True) == None
-    assert repair_json("1.2", return_objects=True) == 1.2
-    assert repair_json("[", return_objects=True) == []
-    assert repair_json("[1, 2, 3, 4", return_objects=True) == [1, 2, 3, 4]
-    assert repair_json("{", return_objects=True) == {}
-    assert repair_json('{ "key": value, "key2": 1 "key3": null }', return_objects=True) == { "key": "value", "key2": 1, "key3": None }
-
-def test_valid_json():
-    assert (
-        repair_json('{"name": "John", "age": 30, "city": "New York"}')
-        == '{"name": "John", "age": 30, "city": "New York"}'
-    )
-    assert (
-        repair_json('{"employees":["John", "Anna", "Peter"]} ')
-        == '{"employees": ["John", "Anna", "Peter"]}'
-    )
-    assert repair_json('{"key": "value:value"}') == '{"key": "value:value"}'
-    assert (
-        repair_json('{"text": "The quick brown fox,"}')
-        == '{"text": "The quick brown fox,"}'
-    )
-    assert (
-        repair_json('{"text": "The quick brown fox won\'t jump"}')
-        == '{"text": "The quick brown fox won\'t jump"}'
-    )
-    assert repair_json('{"key": ""') == '{"key": ""}'
-    assert (
-        repair_json('{"key1": {"key2": [1, 2, 3]}}') == '{"key1": {"key2": [1, 2, 3]}}'
-    )
-    assert (
-        repair_json('{"key": 12345678901234567890}') == '{"key": 12345678901234567890}'
-    )
-    assert repair_json('{"key": "value\u263A"}') == '{"key": "value\\u263a"}'
-    assert repair_json('{"key": "value\\nvalue"}') == '{"key": "value\\nvalue"}'
-
-def test_brackets_edge_cases():
-    assert repair_json("[{]") == "[{}]"
-    assert repair_json("   {  }   ") == "{}"
-    assert repair_json("[") == "[]"
-    assert repair_json("]") == '""'
-    assert repair_json("{") == "{}"
-    assert repair_json("}") == '""'
-    assert repair_json('{"') == '{}'
-    assert repair_json('["') == '[]'
-    assert repair_json('{foo: [}') == '{"foo": []}'
-
-def test_general_edge_cases():
-    assert repair_json("\"") == '""'
-    assert repair_json("\n") == '""'
-    assert repair_json(" ") == '""'
-    assert repair_json("[[1\n\n]") == "[[1]]"
-    assert repair_json("string") == '""'
-    assert repair_json("stringbeforeobject {}") == '{}'
-
-def test_mixed_data_types(): 
-    assert repair_json('  {"key": true, "key2": false, "key3": null}') == '{"key": true, "key2": false, "key3": null}'
-    assert repair_json('{"key": TRUE, "key2": FALSE, "key3": Null}   ') == '{"key": true, "key2": false, "key3": null}'
-
-def test_missing_and_mixed_quotes():
-    assert repair_json("{'key': 'string', 'key2': false, \"key3\": null, \"key4\": unquoted}") == '{"key": "string", "key2": false, "key3": null, "key4": "unquoted"}'
-    assert (
-        repair_json('{"name": "John", "age": 30, "city": "New York')
-        == '{"name": "John", "age": 30, "city": "New York"}'
-    )
-    assert (
-        repair_json('{"name": "John", "age": 30, city: "New York"}')
-        == '{"name": "John", "age": 30, "city": "New York"}'
-    )
-    assert (
-        repair_json('{"name": "John", "age": 30, "city": New York}')
-        == '{"name": "John", "age": 30, "city": "New York"}'
-    )
-    assert (
-        repair_json('{"name": John, "age": 30, "city": "New York"}')
-        == '{"name": "John", "age": 30, "city": "New York"}'
-    )
-    assert repair_json('{“slanted_delimiter”: "value"}') == '{"slanted_delimiter": "value"}'
-    assert (
-        repair_json('{"name": "John", "age": 30, "city": "New')
-        == '{"name": "John", "age": 30, "city": "New"}'
-    )
-    assert repair_json('[{"key": "value", COMMENT "notes": "lorem "ipsum", sic." }]') == '[{"key": "value", "notes": "lorem \\"ipsum\\", sic."}]'
-    assert repair_json('{"key": ""value"}') == '{"key": "value"}'
-    assert repair_json('{"key": "value", 5: "value"}') == '{"key": "value", "5": "value"}'
-    assert repair_json('{"foo": "\\"bar\\""') == '{"foo": "\\"bar\\""}'
-    assert repair_json('{"" key":"val"') == '{" key": "val"}'
-    assert repair_json('{"key": value "key2" : "value2" ') == '{"key": "value", "key2": "value2"}'
-
-def test_array_edge_cases():
-    assert repair_json("[1, 2, 3,") == "[1, 2, 3]"
-    assert repair_json("[1, 2, 3, ...]") == "[1, 2, 3]"
-    assert repair_json("[1, 2, ... , 3]") == "[1, 2, 3]"
-    assert repair_json("[1, 2, '...', 3]") == '[1, 2, "...", 3]'
-    assert repair_json("[true, false, null, ...]") == '[true, false, null]'
-    assert repair_json('["a" "b" "c" 1') == '["a", "b", "c", 1]'
-    assert repair_json('{"employees":["John", "Anna",') == '{"employees": ["John", "Anna"]}'
-    assert repair_json('{"employees":["John", "Anna", "Peter') == '{"employees": ["John", "Anna", "Peter"]}'
-    assert repair_json('{"key1": {"key2": [1, 2, 3') == '{"key1": {"key2": [1, 2, 3]}}'
-    
-def test_escaping():
-    assert repair_json("'\"'") == '""'
-    assert repair_json("{\"key\": 'string\"\n\t\le'") == '{"key": "string\\"\\n\\tle"}'
-    assert repair_json(r'{"real_content": "Some string: Some other string \t Some string <a href=\"https://domain.com\">Some link</a>"') == r'{"real_content": "Some string: Some other string \t Some string <a href=\"https://domain.com\">Some link</a>"}'
-    assert repair_json('{"key_1\n": "value"}') == '{"key_1": "value"}'
-    assert repair_json('{"key\t_": "value"}') == '{"key\\t_": "value"}'
-    
-    
-def test_object_edge_cases():
-    assert repair_json('{       ') == '{}'
-    assert repair_json('{"": "value"') == '{"": "value"}'
-    assert repair_json('{"value_1": true, COMMENT "value_2": "data"}') == '{"value_1": true, "value_2": "data"}'
-    assert repair_json('{"value_1": true, SHOULD_NOT_EXIST "value_2": "data" AAAA }') == '{"value_1": true, "value_2": "data"}'
-    assert repair_json('{"" : true, "key2": "value2"}') == '{"": true, "key2": "value2"}'
-    assert repair_json('{""answer"":[{""traits"":''Female aged 60+'',""answer1"":""5""}]}') == '{"answer": [{"traits": "Female aged 60+", "answer1": "5"}]}'
-    assert repair_json('{ "words": abcdef", "numbers": 12345", "words2": ghijkl" }') == '{"words": "abcdef", "numbers": 12345, "words2": "ghijkl"}'
-    assert repair_json('''{"number": 1,"reason": "According...""ans": "YES"}''') == '{"number": 1, "reason": "According...", "ans": "YES"}'
-    assert repair_json('''{ "a" : "{ b": {} }" }''') == '{"a": "{ b"}'
-    assert repair_json("""{"b": "xxxxx" true}""") == '{"b": "xxxxx"}'
-    assert repair_json('{"key": "Lorem "ipsum" s,"}') == '{"key": "Lorem \\"ipsum\\" s,"}'
-    assert repair_json('{"lorem": ipsum, sic, datum.",}') == '{"lorem": "ipsum, sic, datum."}'
-    assert repair_json('{"lorem": sic tamet. "ipsum": sic tamet, quick brown fox. "sic": ipsum}') == '{"lorem": "sic tamet.", "ipsum": "sic tamet", "sic": "ipsum"}'
-    assert repair_json('{"key":value, " key2":"value2" }') == '{"key": "value", " key2": "value2"}'
-    assert repair_json('{"key":value "key2":"value2" }') == '{"key": "value", "key2": "value2"}'
-
-def test_number_edge_cases():
-    assert repair_json(' - { "test_key": ["test_value", "test_value2"] }') == '{"test_key": ["test_value", "test_value2"]}'
-    assert repair_json('{"key": 1/3}') == '{"key": "1/3"}'
-    assert repair_json('{"key": .25}') == '{"key": 0.25}'
-    assert repair_json('{"here": "now", "key": 1/3, "foo": "bar"}') == '{"here": "now", "key": "1/3", "foo": "bar"}'
-    assert repair_json('{"key": 12345/67890}') == '{"key": "12345/67890"}'
-    assert repair_json('[105,12') == '[105, 12]'
-    assert repair_json('{"key", 105,12,') == '{"key": "105,12"}'
-    assert repair_json('{"key": 1/3, "foo": "bar"}') == '{"key": "1/3", "foo": "bar"}'
-    assert repair_json('{"key": 10-20}') == '{"key": "10-20"}'
-    assert repair_json('{"key": 1.1.1}') == '{"key": "1.1.1"}'
-    assert repair_json('[- ') == '[]'
-
-def test_markdown():
-    assert repair_json('{ "content": "[LINK]("https://google.com")" }') == '{"content": "[LINK](\\"https://google.com\\")"}'
-    assert repair_json('{ "content": "[LINK](" }') == '{"content": "[LINK]("}'
-    assert repair_json('{ "content": "[LINK](", "key": true }') == '{"content": "[LINK](", "key": true}'
-
-def test_leading_trailing_characters():
-    assert repair_json('````{ "key": "value" }```') == '{"key": "value"}'
-    assert repair_json("""{    "a": "",    "b": [ { "c": 1} ] \n}```""") == '{"a": "", "b": [{"c": 1}]}'
-    assert repair_json("Based on the information extracted, here is the filled JSON output: ```json { 'a': 'b' } ```") == '{"a": "b"}'
-    assert repair_json("""
-                       The next 64 elements are:
-                       ```json
-                       { "key": "value" }
-                       ```""") == '{"key": "value"}'
-def test_multiple_jsons():
-    assert repair_json("[]{}") == "[[], {}]"
-    assert repair_json("{}[]{}") == "[{}, [], {}]"
-    assert repair_json('{"key":"value"}[1,2,3,True]') == '[{"key": "value"}, [1, 2, 3, true]]'
-    assert repair_json('lorem ```json {"key":"value"} ``` ipsum ```json [1,2,3,True] ``` 42') == '[{"key": "value"}, [1, 2, 3, true]]'
-
-def test_repair_json_with_objects():
-    # Test with valid JSON strings
-    assert repair_json("[]", return_objects=True) == []
-    assert repair_json("{}", return_objects=True) == {}
-    assert repair_json('{"key": true, "key2": false, "key3": null}', return_objects=True) == {"key": True, "key2": False, "key3": None}
-    assert repair_json('{"name": "John", "age": 30, "city": "New York"}', return_objects=True) == {
-        "name": "John",
-        "age": 30,
-        "city": "New York",
-    }
-    assert repair_json("[1, 2, 3, 4]", return_objects=True) == [1, 2, 3, 4]
-    assert repair_json('{"employees":["John", "Anna", "Peter"]} ', return_objects=True) == {
-        "employees": ["John", "Anna", "Peter"]
-    }
-    assert repair_json('''
-{
-  "resourceType": "Bundle",
-  "id": "1",
-  "type": "collection",
-  "entry": [
-    {
-      "resource": {
-        "resourceType": "Patient",
-        "id": "1",
-        "name": [
-          {"use": "official", "family": "Corwin", "given": ["Keisha", "Sunny"], "prefix": ["Mrs."},
-          {"use": "maiden", "family": "Goodwin", "given": ["Keisha", "Sunny"], "prefix": ["Mrs."]}
-        ]
-      }
-    }
-  ]
-}
-''', return_objects=True) == {"resourceType": "Bundle", "id": "1", "type": "collection", "entry": [{"resource": {"resourceType": "Patient", "id": "1", "name": [{"use": "official", "family": "Corwin", "given": ["Keisha", "Sunny"], "prefix": ["Mrs."]}, {"use": "maiden", "family": "Goodwin", "given": ["Keisha", "Sunny"], "prefix": ["Mrs."]}]}}]}
-    assert repair_json('{\n"html": "<h3 id="aaa">Waarom meer dan 200 Technical Experts - "Passie voor techniek"?</h3>"}', return_objects=True) == {'html': '<h3 id="aaa">Waarom meer dan 200 Technical Experts - "Passie voor techniek"?</h3>'}
-    assert repair_json("""
-        [
-            {
-                "foo": "Foo bar baz",
-                "tag": "#foo-bar-baz"
-            },
-            {
-                "foo": "foo bar "foobar" foo bar baz.",
-                "tag": "#foo-bar-foobar"
-            }
-        ]
-        """, return_objects=True) == [{"foo": "Foo bar baz", "tag": "#foo-bar-baz"},{"foo": "foo bar \"foobar\" foo bar baz.", "tag": "#foo-bar-foobar" }]
-
-def test_repair_json_skip_json_loads():
-    assert repair_json('{"key": true, "key2": false, "key3": null}', skip_json_loads=True) == '{"key": true, "key2": false, "key3": null}'
-    assert repair_json('{"key": true, "key2": false, "key3": null}', return_objects=True, skip_json_loads=True) == {"key": True, "key2": False, "key3": None}
-    assert repair_json('{"key": true, "key2": false, "key3": }', skip_json_loads=True) == '{"key": true, "key2": false, "key3": ""}'
-    assert loads('{"key": true, "key2": false, "key3": }', skip_json_loads=True) == {"key": True, "key2": False, "key3": ""}
-
-
-def test_repair_json_from_file():
-    import os.path
-    import pathlib
-    path = pathlib.Path(__file__).parent.resolve()
-
-    assert from_file(os.path.join(path,"invalid.json")) == [{"_id": "655b66256574f09bdae8abe8", "index": 0, "guid": "31082ae3-b0f3-4406-90f4-cc450bd4379d", "isActive": False, "balance": "$2,562.78", "picture": "http://placehold.it/32x32", "age": 32, "eyeColor": "brown", "name": "Glover Rivas", "gender": "male", "company": "EMPIRICA", "email": "gloverrivas@empirica.com", "phone": "+1 (842) 507-3063", "address": "536 Montague Terrace, Jenkinsville, Kentucky, 2235", "about": "Mollit consectetur excepteur voluptate tempor dolore ullamco enim irure ullamco non enim officia. Voluptate occaecat proident laboris ea Lorem cupidatat reprehenderit nisi nisi aliqua. Amet nulla ipsum deserunt excepteur amet ad aute aute ex. Et enim minim sit veniam est quis dolor nisi sunt quis eiusmod in. Amet eiusmod cillum sunt occaecat dolor laboris voluptate in eiusmod irure aliqua duis.", "registered": "2023-11-18T09:32:36 -01:00", "latitude": 36.26102, "longitude": -91.304608, "tags": ["non", "tempor", "do", "ullamco", "dolore", "sunt", "ipsum"], "friends": [{"id": 0, "name": "Cara Shepherd"}, {"id": 1, "name": "Mason Farley"}, {"id": 2, "name": "Harriet Cochran"}], "greeting": "Hello, Glover Rivas! You have 7 unread messages.", "favoriteFruit": "strawberry"}, {"_id": "655b662585364bc57278bb6f", "index": 1, "guid": "0dea7a3a-f812-4dde-b78d-7a9b58e5da05", "isActive": True, "balance": "$1,359.48", "picture": "http://placehold.it/32x32", "age": 38, "eyeColor": "brown", "name": "Brandi Moreno", "gender": "female", "company": "MARQET", "email": "brandimoreno@marqet.com", "phone": "+1 (850) 434-2077", "address": "537 Doone Court, Waiohinu, Michigan, 3215", "about": "Irure proident adipisicing do Lorem do incididunt in laborum in eiusmod eiusmod ad elit proident. Eiusmod dolor ex magna magna occaecat. Nulla deserunt velit ex exercitation et irure sunt. Cupidatat ut excepteur ea quis labore sint cupidatat incididunt amet eu consectetur cillum ipsum proident. Occaecat exercitation aute laborum dolor proident reprehenderit laborum in voluptate culpa. Exercitation nulla adipisicing culpa aute est deserunt ea nisi deserunt consequat occaecat ut et non. Incididunt ex exercitation dolor dolor anim cillum dolore.", "registered": "2015-09-03T11:47:15 -02:00", "latitude": -19.768953, "longitude": 8.948458, "tags": ["laboris", "occaecat", "laborum", "laborum", "ex", "cillum", "occaecat"], "friends": [{"id": 0, "name": "Erna Kelly"}, {"id": 1, "name": "Black Mays"}, {"id": 2, "name": "Davis Buck"}], "greeting": "Hello, Brandi Moreno! You have 1 unread messages.", "favoriteFruit": "apple"}, {"_id": "655b6625870da431bcf5e0c2", "index": 2, "guid": "b17f6e3f-c898-4334-abbf-05cf222f143b", "isActive": False, "balance": "$1,493.77", "picture": "http://placehold.it/32x32", "age": 20, "eyeColor": "brown", "name": "Moody Meadows", "gender": "male", "company": "OPTIQUE", "email": "moodymeadows@optique.com", "phone": "+1 (993) 566-3041", "address": "766 Osborn Street, Bath, Maine, 7666", "about": "Non commodo excepteur nostrud qui adipisicing aliquip dolor minim nulla culpa proident. In ad cupidatat ea mollit ex est do deserunt proident nostrud. Cillum id id eiusmod amet exercitation nostrud cillum sunt deserunt dolore deserunt eiusmod mollit. Ut ex tempor ad laboris voluptate labore id officia fugiat exercitation amet.", "registered": "2015-01-16T02:48:28 -01:00", "latitude": -25.847327, "longitude": 63.95991, "tags": ["aute", "commodo", "adipisicing", "nostrud", "duis", "mollit", "ut"], "friends": [{"id": 0, "name": "Lacey Cash"}, {"id": 1, "name": "Gabrielle Harmon"}, {"id": 2, "name": "Ellis Lambert"}], "greeting": "Hello, Moody Meadows! You have 4 unread messages.", "favoriteFruit": "strawberry"}, {"_id": "655b6625f3e1bf422220854e", "index": 3, "guid": "92229883-2bfd-4974-a08c-1b506b372e46", "isActive": False, "balance": "$2,215.34", "picture": "http://placehold.it/32x32", "age": 22, "eyeColor": "brown", "name": "Heath Nguyen", "gender": "male", "company": "BLEENDOT", "email": "heathnguyen@bleendot.com", "phone": "+1 (989) 512-2797", "address": "135 Milton Street, Graniteville, Nebraska, 276", "about": "Consequat aliquip irure Lorem cupidatat nulla magna ullamco nulla voluptate adipisicing anim consectetur tempor aliquip. Magna aliqua nulla eu tempor esse proident. Proident fugiat ad ex Lorem reprehenderit dolor aliquip labore labore aliquip. Deserunt aute enim ea minim officia anim culpa sint commodo. Cillum consectetur excepteur aliqua exercitation Lorem veniam voluptate.", "registered": "2016-07-06T01:31:07 -02:00", "latitude": -60.997048, "longitude": -102.397885, "tags": ["do", "ad", "consequat", "irure", "tempor", "elit", "minim"], "friends": [{"id": 0, "name": "Walker Hernandez"}, {"id": 1, "name": "Maria Lane"}, {"id": 2, "name": "Mcknight Barron"}], "greeting": "Hello, Heath Nguyen! You have 4 unread messages.", "favoriteFruit": "apple"}, {"_id": "655b6625519a5b5e4b6742bf", "index": 4, "guid": "c5dc685f-6d0d-4173-b4cf-f5df29a1e8ef", "isActive": True, "balance": "$1,358.90", "picture": "http://placehold.it/32x32", "age": 33, "eyeColor": "brown", "name": "Deidre Duke", "gender": "female", "company": "OATFARM", "email": "deidreduke@oatfarm.com", "phone": "+1 (875) 587-3256", "address": "487 Schaefer Street, Wattsville, West Virginia, 4506", "about": "Laboris eu nulla esse magna sit eu deserunt non est aliqua exercitation commodo. Ad occaecat qui qui laborum dolore anim Lorem. Est qui occaecat irure enim deserunt enim aliqua ex deserunt incididunt esse. Quis in minim laboris proident non mollit. Magna ea do labore commodo. Et elit esse esse occaecat officia ipsum nisi.", "registered": "2021-09-12T04:17:08 -02:00", "latitude": 68.609781, "longitude": -87.509134, "tags": ["mollit", "cupidatat", "irure", "sit", "consequat", "anim", "fugiat"], "friends": [{"id": 0, "name": "Bean Paul"}, {"id": 1, "name": "Cochran Hubbard"}, {"id": 2, "name": "Rodgers Atkinson"}], "greeting": "Hello, Deidre Duke! You have 6 unread messages.", "favoriteFruit": "apple"}, {"_id": "655b6625a19b3f7e5f82f0ea", "index": 5, "guid": "75f3c264-baa1-47a0-b21c-4edac23d9935", "isActive": True, "balance": "$3,554.36", "picture": "http://placehold.it/32x32", "age": 26, "eyeColor": "blue", "name": "Lydia Holland", "gender": "female", "company": "ESCENTA", "email": "lydiaholland@escenta.com", "phone": "+1 (927) 482-3436", "address": "554 Rockaway Parkway, Kohatk, Montana, 6316", "about": "Consectetur ea est labore commodo laborum mollit pariatur non enim. Est dolore et non laboris tempor. Ea incididunt ut adipisicing cillum labore officia tempor eiusmod commodo. Cillum fugiat ex consectetur ut nostrud anim nostrud exercitation ut duis in ea. Eu et id fugiat est duis eiusmod ullamco quis officia minim sint ea nisi in.", "registered": "2018-03-13T01:48:56 -01:00", "latitude": -88.495799, "longitude": 71.840667, "tags": ["veniam", "minim", "consequat", "consequat", "incididunt", "consequat", "elit"], "friends": [{"id": 0, "name": "Debra Massey"}, {"id": 1, "name": "Weiss Savage"}, {"id": 2, "name": "Shannon Guerra"}], "greeting": "Hello, Lydia Holland! You have 5 unread messages.", "favoriteFruit": "banana"}]
-
-    import tempfile
-    # Create a temporary file
-    temp_fd, temp_path = tempfile.mkstemp(suffix=".json")
-    try:
-    # Write content to the temporary file
-        with os.fdopen(temp_fd, 'w') as tmp:
-            tmp.write("{key:value}")
-        assert from_file(temp_path, logging=True) == ({'key': 'value'}, [{'text': 'While parsing a string, we found a literal instead of a quote', 'context': '{key:value}'}, {'text': 'While parsing a string, we found no starting quote. Will add the quote back', 'context': '{key:value}'}, {'context': '{key:value}', 'text': 'While parsing a string missing the left delimiter in object key context, we found a :, stopping here',}, {'text': 'While parsing a string, we missed the closing quote, ignoring', 'context': '{key:value}'}, {'text': 'While parsing a string, we found a literal instead of a quote', 'context': '{key:value}'}, {'text': 'While parsing a string, we found no starting quote. Will add the quote back', 'context': '{key:value}'}, {'context': '{key:value}', 'text': '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'}, {'text': 'While parsing a string, we missed the closing quote, ignoring', 'context': '{key:value}'}])
-    finally:
-        # Clean up - delete the temporary file
-        os.remove(temp_path)
-
-def test_ensure_ascii():
-    assert repair_json("{'test_中国人_ascii':'统一码'}", ensure_ascii=False) == '{"test_中国人_ascii": "统一码"}'