Repository navigation
Commit b3329ab
committed
[fix.perception.rho.D8+] 1GRM offered its entire 300 KB mmCIF file as a ligand name
The ligand selector's first option on 1GRM was the whole deposited file — 300,821 characters
— and it travelled from there into the ligand panel header and the properties cockpit header,
so the app rendered a crystallographic file where a residue name belongs.
The cause is not a bad label, it is an INVALID LOCATION. StructureProperties.residue.key came
back undefined for that entry, and chain.auth_asym_id then returned the raw file. Every
property read off that location is garbage rather than merely wrong.
What the first attempt got wrong, and the reason it is written down: I guarded `comp`, and
the result was "?undefined · <the whole file>" — the corruption simply moved to the next
field. Guarding a symptom field by field chases it around the struct; asking whether the
LOCATION is valid answers it once. A location that cannot yield a residue key is not
addressing a residue, so the group does not enter the list at all.
Verified: 1GRM now offers FVA · A:1 and FVA · B:1 (9 atoms each) and the cockpit reads
MW 129 · LogP -0.0, and the deposited-ligand structures are unchanged.1 parent c7e7cea commit b3329ab
1 file changed
Lines changed: 15 additions & 1 deletion
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
125 | 125 | | |
126 | 126 | | |
127 | 127 | | |
| 128 | + | |
| 129 | + | |
| 130 | + | |
| 131 | + | |
| 132 | + | |
| 133 | + | |
| 134 | + | |
| 135 | + | |
| 136 | + | |
| 137 | + | |
| 138 | + | |
| 139 | + | |
128 | 140 | | |
129 | 141 | | |
130 | 142 | | |
131 | 143 | | |
132 | 144 | | |
133 | 145 | | |
134 | | - | |
| 146 | + | |
| 147 | + | |
| 148 | + | |
135 | 149 | | |
136 | 150 | | |
137 | 151 | | |
| |||
0 commit comments