Fix bug where black tries to split string on escaped space (#1799)

Closes #1505.

diff --git a/src/black/__init__.py b/src/black/__init__.py
index 24e9d4edaaaeb5fa8792e9f0c21f0e9e863d1047..e09838d866a1f3071079858fc12afc70575cdff8 100644 (file)
--- a/src/black/__init__.py
+++ b/src/black/__init__.py
@@ -3590,7 +3590,8 @@ class StringSplitter(CustomSplitMapMixin, BaseStringSplitter):
         MIN_SUBSTR_SIZE characters.
 
         The string will ONLY be split on spaces (i.e. each new substring should
         MIN_SUBSTR_SIZE characters.
 
         The string will ONLY be split on spaces (i.e. each new substring should

-        start with a space).
+        start with a space). Note that the string will NOT be split on a space
+        which is escaped with a backslash.

 
         If the string is an f-string, it will NOT be split in the middle of an
         f-expression (e.g. in f"FooBar: {foo() if x else bar()}", {foo() if x
 
         If the string is an f-string, it will NOT be split in the middle of an
         f-expression (e.g. in f"FooBar: {foo() if x else bar()}", {foo() if x


@@ -3930,11 +3931,23 @@ class StringSplitter(CustomSplitMapMixin, BaseStringSplitter):
                 section of this classes' docstring would be be met by returning @i.
             """
             is_space = string[i] == " "
                 section of this classes' docstring would be be met by returning @i.
             """
             is_space = string[i] == " "

+
+            is_not_escaped = True
+            j = i - 1
+            while is_valid_index(j) and string[j] == "\\":
+                is_not_escaped = not is_not_escaped
+                j -= 1
+

             is_big_enough = (
                 len(string[i:]) >= self.MIN_SUBSTR_SIZE
                 and len(string[:i]) >= self.MIN_SUBSTR_SIZE
             )
             is_big_enough = (
                 len(string[i:]) >= self.MIN_SUBSTR_SIZE
                 and len(string[:i]) >= self.MIN_SUBSTR_SIZE
             )

-            return is_space and is_big_enough and not breaks_fstring_expression(i)
+            return (
+                is_space
+                and is_not_escaped
+                and is_big_enough
+                and not breaks_fstring_expression(i)
+            )

 
         # First, we check all indices BELOW @max_break_idx.
         break_idx = max_break_idx
 
         # First, we check all indices BELOW @max_break_idx.
         break_idx = max_break_idx

diff --git a/tests/data/long_strings__regression.py b/tests/data/long_strings__regression.py
index 8290a4cbc1c1c161bfd27494ae0c8edfd91fc07a..9bfb1eedd5aaef71c73c5c19790b61e014b12d54 100644 (file)
--- a/tests/data/long_strings__regression.py
+++ b/tests/data/long_strings__regression.py
@@ -353,6 +353,24 @@ value.__dict__[
     key
 ] = "test"  # set some Thrift field to non-None in the struct aa bb cc dd ee
 
     key
 ] = "test"  # set some Thrift field to non-None in the struct aa bb cc dd ee
 

+RE_ONE_BACKSLASH = {
+    "asdf_hjkl_jkl": re.compile(
+        r"(?<!([0-9]\ ))(?<=(^|\ ))([A-Z]+(\ )?|[0-9](\ )|[a-z](\ )){4,7}([A-Z]|[0-9]|[a-z])($|\b)(?!(\ ?([0-9]\ )|(\.)))"
+    ),
+}
+
+RE_TWO_BACKSLASHES = {
+    "asdf_hjkl_jkl": re.compile(
+        r"(?<!([0-9]\ ))(?<=(^|\ ))([A-Z]+(\ )?|[0-9](\ )|[a-z](\\ )){4,7}([A-Z]|[0-9]|[a-z])($|\b)(?!(\ ?([0-9]\ )|(\.)))"
+    ),
+}
+
+RE_THREE_BACKSLASHES = {
+    "asdf_hjkl_jkl": re.compile(
+        r"(?<!([0-9]\ ))(?<=(^|\ ))([A-Z]+(\ )?|[0-9](\ )|[a-z](\\\ )){4,7}([A-Z]|[0-9]|[a-z])($|\b)(?!(\ ?([0-9]\ )|(\.)))"
+    ),
+}
+

 # output
 
 
 # output
 
 


@@ -793,3 +811,22 @@ class xxxxxxxxxxxxxxxxxxxxx(xxxx.xxxxxxxxxxxxx):
 value.__dict__[
     key
 ] = "test"  # set some Thrift field to non-None in the struct aa bb cc dd ee
 value.__dict__[
     key
 ] = "test"  # set some Thrift field to non-None in the struct aa bb cc dd ee

+
+RE_ONE_BACKSLASH = {
+    "asdf_hjkl_jkl": re.compile(
+        r"(?<!([0-9]\ ))(?<=(^|\ ))([A-Z]+(\ )?|[0-9](\ )|[a-z](\ )){4,7}([A-Z]|[0-9]|[a-z])($|\b)(?!(\ ?([0-9]\ )|(\.)))"
+    ),
+}
+
+RE_TWO_BACKSLASHES = {
+    "asdf_hjkl_jkl": re.compile(
+        r"(?<!([0-9]\ ))(?<=(^|\ ))([A-Z]+(\ )?|[0-9](\ )|[a-z](\\"
+        r" )){4,7}([A-Z]|[0-9]|[a-z])($|\b)(?!(\ ?([0-9]\ )|(\.)))"
+    ),
+}
+
+RE_THREE_BACKSLASHES = {
+    "asdf_hjkl_jkl": re.compile(
+        r"(?<!([0-9]\ ))(?<=(^|\ ))([A-Z]+(\ )?|[0-9](\ )|[a-z](\\\ )){4,7}([A-Z]|[0-9]|[a-z])($|\b)(?!(\ ?([0-9]\ )|(\.)))"
+    ),
+}