daniel/migrants
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Fix IRI generation by sanitizing column names with spaces.

Browse source 
Regenerate graph-01.ttl and graph-02.ttl with the corrected output.
This commit is contained in:
Daniel Hernandez 2026-02-28 06:20:18 +01:00
parent 34754cc68a
commit aa6f46bfc2
3 changed files with 326756 additions and 367995 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

344115
data/graph-01.ttl
View file

File diff suppressed because it is too large Load diff

350623
data/graph-02.ttl
View file

File diff suppressed because it is too large Load diff

13
src/map/step_01.rs
View file

@ -79,10 +79,19 @@ fn class_iri(table: &str) -> SimpleTerm<'static> {
)
}
fn sanitize_name(name: &str) -> String {
let s: String = name
.chars()
.map(|c| if c.is_ascii_alphanumeric() || c == '-' { c } else { '_' })
.collect();
let s = s.replace("__", "_");
s.trim_matches('_').to_string()
}
fn column_iri(table: &str, column: &str) -> SimpleTerm<'static> {
SimpleTerm::Iri(
sophia::api::term::IriRef::new_unchecked(
format!("{}{}#{}", BASE_IRI, table, column).into(),
format!("{}{}#{}", BASE_IRI, table, sanitize_name(column)).into(),
),
)
}
@ -90,7 +99,7 @@ fn column_iri(table: &str, column: &str) -> SimpleTerm<'static> {
fn ref_iri(table: &str, fk_col: &str) -> SimpleTerm<'static> {
SimpleTerm::Iri(
sophia::api::term::IriRef::new_unchecked(
format!("{}{}#ref-{}", BASE_IRI, table, fk_col).into(),
format!("{}{}#ref-{}", BASE_IRI, table, sanitize_name(fk_col)).into(),
),
)
}