diff options
Diffstat (limited to 'src/sisudoc/ocda/io_in/read_zip_pod.d')
| -rw-r--r-- | src/sisudoc/ocda/io_in/read_zip_pod.d | 26 |
1 files changed, 23 insertions, 3 deletions
diff --git a/src/sisudoc/ocda/io_in/read_zip_pod.d b/src/sisudoc/ocda/io_in/read_zip_pod.d index 49429be..d0eb65a 100644 --- a/src/sisudoc/ocda/io_in/read_zip_pod.d +++ b/src/sisudoc/ocda/io_in/read_zip_pod.d @@ -63,6 +63,7 @@ template spineExtractZipPod() { import std.regex; import std.stdio; import std.string : indexOf; + import sisudoc.ocda.zstd; /+ security limits for zip extraction +/ enum size_t MAX_ENTRY_SIZE = 50 * 1024 * 1024; /+ 50 MB per entry +/ @@ -122,10 +123,29 @@ template spineExtractZipPod() { } /+ ↓ derive pod name from zip filename +/ string zip_basename = zip_path.baseName.stripExtension; - /+ ↓ read and parse zip archive +/ + /+ ↓ read and parse the pod archive. + a .sisupod is one zstd frame wrapping the archive spine writes, so + the bytes are unwrapped before the archive is parsed. Determined from + the first four bytes and not from the filename: a pod published as a + plain .zip, or one a reader has renamed, is read either way. Everything + after this point sees the same archive it always did, guards included. + +/ + ubyte[] _archive_bytes; + try { + _archive_bytes = cast(ubyte[]) read(zip_path); + if (zstdIsFrame(_archive_bytes)) { + _archive_bytes = zstdDecompress(_archive_bytes, MAX_TOTAL_SIZE); + } + } catch (ZstdException ex) { + result.error_msg = "failed to decompress pod archive: " ~ zip_path ~ " - " ~ ex.msg; + return result; + } catch (Exception ex) { + result.error_msg = "error reading pod archive: " ~ zip_path ~ " - " ~ ex.msg; + return result; + } ZipArchive zip; try { - zip = new ZipArchive(read(zip_path)); + zip = new ZipArchive(_archive_bytes); } catch (ZipException ex) { result.error_msg = "failed to read zip archive: " ~ zip_path ~ " - " ~ ex.msg; return result; @@ -274,7 +294,7 @@ template spineExtractZipPod() { its own, carrying its images. A .ssp is not here: it describes its images without carrying them, so it would arrive without them. +/ - static auto rgx_url_source = ctRegex!(`^https?://[a-zA-Z0-9._:/-]+([.]zip|[.]ocda[.]db)$`); + static auto rgx_url_source = ctRegex!(`^https?://[a-zA-Z0-9._:/-]+([.]sisupod|[.]zip|[.]ocda[.]db)$`); struct DownloadResult { string local_path; /+ path to downloaded temp file +/ |
