Skip to content

Commit

Permalink
fix regex anchoring semantics (#82)
Browse files Browse the repository at this point in the history 
From json-schema.org:
When defining the regular expressions, it's important to note that the string is considered valid if the expression matches anywhere within the string.
  • Loading branch information
@hudson-ai
hudson-ai authored Dec 10, 2024
1 parent a7b69d6 commit 904562a
Show file tree
Hide file tree
Showing 3 changed files with 26 additions and 14 deletions.
18 changes: 15 additions & 3 deletions parser/src/json/compiler.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -563,9 +563,21 @@ impl Compiler {
if min_length > 0 || max_length.is_some() {
bail!("If a pattern is specified, minLength and maxLength must be unspecified.");
}
// the regex has implicit ^...$ anyways
let regex = regex.trim_start_matches('^').trim_end_matches('$');
let node = self.builder.lexeme(mk_regex(regex), true);
let regex = {
let left_anchored = regex.starts_with('^');
let right_anchored = regex.ends_with('$');
let trimmed = regex.trim_start_matches('^').trim_end_matches('$');
let mut result = String::new();
if !left_anchored {
result.push_str(".*");
}
result.push_str(trimmed);
if !right_anchored {
result.push_str(".*");
}
result
};
let node = self.builder.lexeme(mk_regex(&regex), true);
Ok(node)
} else {
Ok(self.lexeme(&format!(
Expand Down
20 changes: 10 additions & 10 deletions parser/src/json/formats.rs
View file
Original file line number Diff line number Diff line change
@@ -1,31 +1,31 @@
pub fn lookup_format(name: &str) -> Option<&str> {
let r = match name {
"date-time" => {
r"(?P<date>[0-9]{4}-(?:0[1-9]|1[0-2])-(?:0[1-9]|[12][0-9]|3[01]))[tT](?P<time>(?:[01][0-9]|2[0-3]):[0-5][0-9]:(?:[0-5][0-9]|60)(?P<time_fraction>\.[0-9]+)?(?P<time_zone>[zZ]|[+-](?:[01][0-9]|2[0-3]):[0-5][0-9]))"
r"^(?P<date>[0-9]{4}-(?:0[1-9]|1[0-2])-(?:0[1-9]|[12][0-9]|3[01]))[tT](?P<time>(?:[01][0-9]|2[0-3]):[0-5][0-9]:(?:[0-5][0-9]|60)(?P<time_fraction>\.[0-9]+)?(?P<time_zone>[zZ]|[+-](?:[01][0-9]|2[0-3]):[0-5][0-9]))$"
}
"time" => {
r"(?:[01][0-9]|2[0-3]):[0-5][0-9]:(?:[0-5][0-9]|60)(?P<time_fraction>\.[0-9]+)?(?P<time_zone>[zZ]|[+-](?:[01][0-9]|2[0-3]):[0-5][0-9])"
r"^(?:[01][0-9]|2[0-3]):[0-5][0-9]:(?:[0-5][0-9]|60)(?P<time_fraction>\.[0-9]+)?(?P<time_zone>[zZ]|[+-](?:[01][0-9]|2[0-3]):[0-5][0-9])$"
}
"date" => r"[0-9]{4}-(?:0[1-9]|1[0-2])-(?:0[1-9]|[12][0-9]|3[01])",
"date" => r"^[0-9]{4}-(?:0[1-9]|1[0-2])-(?:0[1-9]|[12][0-9]|3[01])$",
"duration" => {
r"P(?:(?P<dur_date>(?:(?P<dur_year>[0-9]+Y(?:[0-9]+M(?:[0-9]+D)?)?)|(?P<dur_month>[0-9]+M(?:[0-9]+D)?)|(?P<dur_day>[0-9]+D))(?:T(?:(?P<dur_hour>[0-9]+H(?:[0-9]+M(?:[0-9]+S)?)?)|(?P<dur_minute>[0-9]+M(?:[0-9]+S)?)|(?P<dur_second>[0-9]+S)))?)|(?P<dur_time>T(?:(?P<dur_hour2>[0-9]+H(?:[0-9]+M(?:[0-9]+S)?)?)|(?P<dur_minute2>[0-9]+M(?:[0-9]+S)?)|(?P<dur_second2>[0-9]+S)))|(?P<dur_week>[0-9]+W))"
r"^P(?:(?P<dur_date>(?:(?P<dur_year>[0-9]+Y(?:[0-9]+M(?:[0-9]+D)?)?)|(?P<dur_month>[0-9]+M(?:[0-9]+D)?)|(?P<dur_day>[0-9]+D))(?:T(?:(?P<dur_hour>[0-9]+H(?:[0-9]+M(?:[0-9]+S)?)?)|(?P<dur_minute>[0-9]+M(?:[0-9]+S)?)|(?P<dur_second>[0-9]+S)))?)|(?P<dur_time>T(?:(?P<dur_hour2>[0-9]+H(?:[0-9]+M(?:[0-9]+S)?)?)|(?P<dur_minute2>[0-9]+M(?:[0-9]+S)?)|(?P<dur_second2>[0-9]+S)))|(?P<dur_week>[0-9]+W))$"
}
"email" => {
r"(?P<local_part>(?P<dot_string>[^\s@\.]+(\.[^\s@\.]+)*))@((?P<domain>(?P<sub_domain>[a-zA-Z0-9]([a-zA-Z0-9-]*[a-zA-Z0-9])?)(\.(?P<sub_domain2>[a-zA-Z0-9]([a-zA-Z0-9-]*[a-zA-Z0-9])?))*)|\[(?P<ipv4>((([0-9])|(([1-9])[0-9]|(25[0-5]|(2[0-4]|(1)[0-9])[0-9])))\.){3}(([0-9])|(([1-9])[0-9]|(25[0-5]|(2[0-4]|(1)[0-9])[0-9]))))\])"
r"^(?P<local_part>(?P<dot_string>[^\s@\.]+(\.[^\s@\.]+)*))@((?P<domain>(?P<sub_domain>[a-zA-Z0-9]([a-zA-Z0-9-]*[a-zA-Z0-9])?)(\.(?P<sub_domain2>[a-zA-Z0-9]([a-zA-Z0-9-]*[a-zA-Z0-9])?))*)|\[(?P<ipv4>((([0-9])|(([1-9])[0-9]|(25[0-5]|(2[0-4]|(1)[0-9])[0-9])))\.){3}(([0-9])|(([1-9])[0-9]|(25[0-5]|(2[0-4]|(1)[0-9])[0-9]))))\])$"
}
"hostname" => {
r"[a-zA-Z0-9]([a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(\.[a-zA-Z0-9]([a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*"
r"^[a-zA-Z0-9]([a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(\.[a-zA-Z0-9]([a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*$"
}
"ipv4" => {
r"((([0-9])|(([1-9])[0-9]|(25[0-5]|(2[0-4]|(1)[0-9])[0-9])))\.){3}(([0-9])|(([1-9])[0-9]|(25[0-5]|(2[0-4]|(1)[0-9])[0-9])))"
r"^((([0-9])|(([1-9])[0-9]|(25[0-5]|(2[0-4]|(1)[0-9])[0-9])))\.){3}(([0-9])|(([1-9])[0-9]|(25[0-5]|(2[0-4]|(1)[0-9])[0-9])))$"
}
"ipv6" => {
r"(?:(?P<full>(?:[0-9a-fA-F]{1,4}:){7}[0-9a-fA-F]{1,4}))|(?:::(?:[0-9a-fA-F]{1,4}:){0,5}(?P<ls32>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:(?P<h16_1>[0-9a-fA-F]{1,4})?::(?:[0-9a-fA-F]{1,4}:){0,4}(?P<ls32_1>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,1}[0-9a-fA-F]{1,4})?::(?:[0-9a-fA-F]{1,4}:){0,3}(?P<ls32_2>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,2}[0-9a-fA-F]{1,4})?::(?:[0-9a-fA-F]{1,4}:){0,2}(?P<ls32_3>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,3}[0-9a-fA-F]{1,4})?::[0-9a-fA-F]{1,4}:(?P<ls32_4>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,4}[0-9a-fA-F]{1,4})?::(?P<ls32_5>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,5}[0-9a-fA-F]{1,4})?::(?P<h16_2>[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,6}[0-9a-fA-F]{1,4})?::)"
r"^(?:(?P<full>(?:[0-9a-fA-F]{1,4}:){7}[0-9a-fA-F]{1,4}))|(?:::(?:[0-9a-fA-F]{1,4}:){0,5}(?P<ls32>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:(?P<h16_1>[0-9a-fA-F]{1,4})?::(?:[0-9a-fA-F]{1,4}:){0,4}(?P<ls32_1>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,1}[0-9a-fA-F]{1,4})?::(?:[0-9a-fA-F]{1,4}:){0,3}(?P<ls32_2>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,2}[0-9a-fA-F]{1,4})?::(?:[0-9a-fA-F]{1,4}:){0,2}(?P<ls32_3>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,3}[0-9a-fA-F]{1,4})?::[0-9a-fA-F]{1,4}:(?P<ls32_4>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,4}[0-9a-fA-F]{1,4})?::(?P<ls32_5>[0-9a-fA-F]{1,4}:[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,5}[0-9a-fA-F]{1,4})?::(?P<h16_2>[0-9a-fA-F]{1,4}))|(?:((?:[0-9a-fA-F]{1,4}:){0,6}[0-9a-fA-F]{1,4})?::)$"
}
"uuid" => {
r"(?P<time_low>[0-9a-fA-F]{8})-(?P<time_mid>[0-9a-fA-F]{4})-(?P<time_high_and_version>[0-9a-fA-F]{4})-(?P<clock_seq_and_reserved>[0-9a-fA-F]{2})(?P<clock_seq_low>[0-9a-fA-F]{2})-(?P<node>[0-9a-fA-F]{12})"
r"^(?P<time_low>[0-9a-fA-F]{8})-(?P<time_mid>[0-9a-fA-F]{4})-(?P<time_high_and_version>[0-9a-fA-F]{4})-(?P<clock_seq_and_reserved>[0-9a-fA-F]{2})(?P<clock_seq_low>[0-9a-fA-F]{2})-(?P<node>[0-9a-fA-F]{12})$"
}
"unknown" => r"(?s:.*)",
"unknown" => r"^(?s:.*)$",
_ => return None,
};
Some(r)
Expand Down
2 changes: 1 addition & 1 deletion parser/src/json/schema.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -554,7 +554,7 @@ fn compile_const(instance: &Value) -> Result<Schema> {
Value::String(s) => Ok(Schema::String {
min_length: 0,
max_length: None,
pattern: Some(escape(s)),
pattern: Some(format!("^{}$", escape(s))),
format: None,
}),
Value::Array(items) => {
Expand Down

0 comments on commit 904562a

Please sign in to comment.