///|
/// Checks the properties that only make sense across sentences: every sentence
/// carrying a `# sent_id`, no repeated `# sent_id`, and no document or paragraph
/// id reused inside the same scope. Sentence-level structure is already covered
/// by `validate_sentence`; these findings are about the file as a corpus.
pub fn validate_documents(sentences : Array[Sentence]) -> Array[Diagnostic] {
let out : Array[Diagnostic] = []
let sent_ids : Array[String] = []
let doc_ids : Array[String] = []
let par_scopes : Array[Int32] = []
let par_ids : Array[String] = []
let mut doc_index : Int32 = -1
for sentence in sentences {
if sentence.has_newdoc() {
doc_index = doc_index + 1
match sentence.newdoc_id() {
Some(id) =>
if doc_ids.contains(id) {
push(
out,
sentence.line,
Warning,
"NEWDOC_ID_DUPLICATE",
"document id \{id} is used by more than one # newdoc",
)
} else {
doc_ids.push(id)
}
None => ()
}
}
let scope = if doc_index < 0 { 0 } else { doc_index }
if sentence.has_newpar() {
if doc_index < 0 {
push(
out,
sentence.line,
Warning,
"NEWPAR_WITHOUT_DOC",
"# newpar appears before any # newdoc",
)
}
match sentence.newpar_id() {
Some(id) => {
let mut seen = false
let mut i = 0
while i < par_ids.length() {
if par_scopes.at(i) == scope && par_ids.at(i) == id {
seen = true
}
i = i + 1
}
if seen {
push(
out,
sentence.line,
Warning,
"NEWPAR_ID_DUPLICATE",
"paragraph id \{id} is reused inside document \{scope}",
)
} else {
par_scopes.push(scope)
par_ids.push(id)
}
}
None => ()
}
}
match sentence.comment_value("sent_id") {
Some(id) =>
if sent_ids.contains(id) {
push(
out,
sentence.line,
Error,
"SENT_ID_DUPLICATE",
"sent_id \{id} is used by more than one sentence",
)
} else {
sent_ids.push(id)
}
None =>
push(
out,
sentence.line,
Warning,
"SENT_ID_MISSING",
"sentence has no # sent_id comment",
)
}
}
out
}