Merge branch 'main' into fix/improve-exception-handling-splitter

Author: MiWeiss

bibtexparser/entrypoint.py

@@ -1,3 +1,4 @@
+import codecs
 import warnings
 from typing import Iterable
 from typing import List
@@ -20,9 +21,9 @@ def _build_parse_stack(
 ) -> List[Middleware]:
     if parse_stack is not None and append_middleware is not None:
         raise ValueError(
-            "Provided both parse_stack and append_middleware."
-            "Only one should be provided."
-            "(append_middleware should only be used with the default parse_stack,"
+            "Provided both parse_stack and append_middleware. "
+            "Only one should be provided. "
+            "(append_middleware should only be used with the default parse_stack, "
             "i.e., when the passed parse_stack is None.)"
         )
 
@@ -50,10 +51,10 @@ def _build_unparse_stack(
 ) -> List[Middleware]:
     if unparse_stack is not None and prepend_middleware is not None:
         raise ValueError(
-            "Provided both parse_stack and append_middleware."
-            "Only one should be provided."
-            "(prepend_middleware should only be used with the default parse_stack,"
-            "i.e., when the passed parse_stack is None.)"
+            "Provided both unparse_stack and prepend_middleware. "
+            "Only one should be provided. "
+            "(prepend_middleware should only be used with the default unparse_stack, "
+            "i.e., when the passed unparse_stack is None.)"
         )
 
     if unparse_stack is None:
@@ -125,7 +126,13 @@ def parse_file(
 
     :param encoding: Encoding of the .bib file. Default encoding is ``"UTF-8"``.
     :return: Library: Parsed BibTeX library
+    :raises LookupError: If the specified encoding is not recognized.
     """
+    try:
+        codecs.lookup(encoding)
+    except LookupError:
+        raise LookupError(f"Unknown encoding: {encoding!r}")
+
     with open(path, encoding=encoding) as f:
         bibtex_str = f.read()
         return parse_string(

bibtexparser/library.py

@@ -107,18 +107,22 @@ def replace(self, old_block: Block, new_block: Block, fail_on_duplicate_key: boo
     def _cast_to_duplicate(
         prev_block_with_same_key: Union[Entry, String], duplicate: Union[Entry, String]
     ):
-        assert isinstance(prev_block_with_same_key, type(duplicate)) or isinstance(
-            duplicate, type(prev_block_with_same_key)
-        ), (
-            "Internal BibtexParser Error. Duplicate blocks share no common type."
-            f"Found {type(prev_block_with_same_key)} and {type(duplicate)}, but both should be"
-            f"either instance of String or instance of Entry."
-            f"Please report this issue at the bibtexparser issue tracker.",
-        )
-
-        assert (
-            prev_block_with_same_key.key == duplicate.key
-        ), "Internal BibtexParser Error. Duplicate blocks have different keys."
+        if not (
+            isinstance(prev_block_with_same_key, type(duplicate))
+            or isinstance(duplicate, type(prev_block_with_same_key))
+        ):
+            raise ValueError(
+                "Internal BibtexParser Error. Duplicate blocks share no common type. "
+                f"Found {type(prev_block_with_same_key)} and {type(duplicate)}, but both should be "
+                "either instance of String or instance of Entry. "
+                "Please report this issue at the bibtexparser issue tracker."
+            )
+
+        if prev_block_with_same_key.key != duplicate.key:
+            raise ValueError(
+                "Internal BibtexParser Error. Duplicate blocks have different keys. "
+                "Please report this issue at the bibtexparser issue tracker."
+            )
 
         return DuplicateBlockKeyBlock(
             start_line=duplicate.start_line,

bibtexparser/middlewares/latex_encoding.py

@@ -102,10 +102,15 @@ def __init__(
 
         if encoder is not None and (keep_math is not None or enclose_urls is not None):
             raise ValueError(
-                "Cannot specify both encoder and keep_math or enclose_urls."
+                "Cannot specify both encoder and keep_math or enclose_urls. "
                 "If you want to use a custom encoder, you have to specify it completely."
             )
 
+        if encoder is not None and not isinstance(encoder, UnicodeToLatexEncoder):
+            raise TypeError(
+                f"encoder must be a UnicodeToLatexEncoder instance, got {type(encoder).__name__}"
+            )
+
         # Defaults (not specified as defaults in args,
         #   to make sure we can identify if they were specified)
         keep_math = keep_math if keep_math is not None else True
@@ -167,11 +172,16 @@ def __init__(
         if decoder is not None and (keep_braced_groups is not None or keep_math_mode is not None):
             raise ValueError(
                 "Cannot specify both decoder and one of "
-                "`keep_braced_groups` or `keep_math_mode`."
+                "`keep_braced_groups` or `keep_math_mode`. "
                 "If you want to use a custom decoder, "
                 "you have to specify it completely."
             )
 
+        if decoder is not None and not isinstance(decoder, LatexNodes2Text):
+            raise TypeError(
+                f"decoder must be a LatexNodes2Text instance, got {type(decoder).__name__}"
+            )
+
         # Defaults (not specified as defaults in args,
         #   to make sure we can identify if they were specified)
         keep_braced_groups = keep_braced_groups if keep_braced_groups is not None else False

bibtexparser/middlewares/names.py

@@ -164,8 +164,8 @@ def _transform_field_value(self, name) -> List[NameParts]:
         if not isinstance(name, list):
             raise ValueError(
                 "Expected a list of strings, got {}. "
-                "Make sure to use `SeparateCoAuthors` middleware"
-                "before using `SplitNameParts` middleware".format(name)
+                "Make sure to use `SeparateCoAuthors` middleware "
+                "before using `SplitNameParts` middleware.".format(name)
             )
 
         return [parse_single_name_into_parts(n) for n in name]