frozenleaves/alphafold3
1
0
Fork 0
mirror of https://github.com/google-deepmind/alphafold3.git synced 2025-10-20 13:23:47 +08:00
Code Issues Packages Projects Releases Wiki Activity

Improve validation of CIF files.

Browse Source 
* Make sure the data name in `data_<name>` is non-empty.
* Check for duplicate key names.
* Check that the multi-line tokens are closed at the end of the file.
* Fix also two broken mmCIF files this check has uncovered.

PiperOrigin-RevId: 802572236
Change-Id: Ie7a3a5ec816ec5b97158508cc1b12064cf0e70a8
This commit is contained in:
Augustin Zidek
2025-09-03 08:50:25 -07:00
committed by Copybara-Service
parent 4208665547
commit b467f92160
1 changed files with 19 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
src/alphafold3/parsers/cpp/cif_dict_lib.cc
View File

@ -134,6 +134,9 @@ absl::StatusOr<std::vector<absl::string_view>> TokenizeInternal(
line_num++; line_num++;
if (!multiline.empty() && multiline[0] == ';') { if (!multiline.empty() && multiline[0] == ';') {
break; break;
} else if (line_num == lines.size()) {
return absl::InvalidArgumentError(
"Last multiline token is not terminated by a semicolon.");
} }
multiline_tokens.push_back(multiline); multiline_tokens.push_back(multiline);
} }
@ -374,6 +377,10 @@ absl::StatusOr<CifDict> CifDict::FromString(absl::string_view cif_string) {
return absl::InvalidArgumentError( return absl::InvalidArgumentError(
"The CIF file does not start with the data_ field."); "The CIF file does not start with the data_ field.");
} }
if (first_token.empty()) {
return absl::InvalidArgumentError(
"The CIF file does not contain a data block name.");
}
cif["data_"].emplace_back(first_token); cif["data_"].emplace_back(first_token);
// Counters for CIF loop_ regions. // Counters for CIF loop_ regions.
@ -413,7 +420,12 @@ absl::StatusOr<CifDict> CifDict::FromString(absl::string_view cif_string) {
loop_flag = false; loop_flag = false;
} else { } else {
// We are in the keys (column names) section of the loop. // We are in the keys (column names) section of the loop.
auto& columns = cif[token]; auto [it, inserted] = cif.try_emplace(token);
if (!inserted) {
return absl::InvalidArgumentError(
absl::StrCat("Duplicate loop key: '", token, "'"));
}
auto& columns = it->second;
columns.clear(); columns.clear();
// Heuristic: _atom_site is typically the largest table in an mmCIF // Heuristic: _atom_site is typically the largest table in an mmCIF
@ -448,7 +460,12 @@ absl::StatusOr<CifDict> CifDict::FromString(absl::string_view cif_string) {
absl::StrCat("Key '", key, "' does not start with an underscore.")); absl::StrCat("Key '", key, "' does not start with an underscore."));
} }
} else { } else {
cif[key].emplace_back(token); auto [it, inserted] = cif.try_emplace(key);
if (!inserted) {
return absl::InvalidArgumentError(
absl::StrCat("Duplicate key: '", key, "'"));
}
(it->second).emplace_back(token);
key = ""; key = "";
} }
} }