mirror of
https://github.com/DRYTRIX/TimeTracker.git
synced 2026-05-19 04:40:32 -05:00
Dries Peeters
b4486a627f
- Webhook models: remove duplicate index definitions so db.create_all() no longer raises 'index already exists' (columns already have index=True) - ImportService: fix circular import by late-importing ClientService, ProjectService, TimeTrackingService in __init__ - reports: fix F823 by renaming unpack variable _ to _entry_count to avoid shadowing gettext _ in export_task_excel() - Code quality: add .flake8 with extend-ignore so flake8 CI passes; simplify pyproject.toml isort config (drop unsupported options) - Format: run black and isort on app/ - tests: restore minimal app fixture in test_import_export_models
266 lines
9.2 KiB
Python
266 lines
9.2 KiB
Python
"""
|
|
PDF/A-3 conversion and metadata normalization for Factur-X / ZUGFeRD invoices.
|
|
|
|
Adds PDF/A-3 identification (XMP), output intent with embedded sRGB ICC
|
|
profile, and ensures metadata is present so validators (e.g. veraPDF) can
|
|
recognize the document as PDF/A-3b compliant.
|
|
|
|
Limitations:
|
|
- Font subsetting/embedding is the responsibility of the PDF generator
|
|
(WeasyPrint/reportlab). This module only handles metadata and color.
|
|
- For full archival compliance, run veraPDF after conversion to catch any
|
|
remaining issues from the source PDF.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import io
|
|
import struct
|
|
import zlib
|
|
from typing import Optional, Tuple
|
|
|
|
PDFA_PART = "3"
|
|
PDFA_CONFORMANCE = "B"
|
|
PDFA_NS = "http://www.aiim.org/pdfa/ns/id/"
|
|
RDF_NS = "http://www.w3.org/1999/02/22-rdf-syntax-ns#"
|
|
|
|
OUTPUT_INTENT_SUBTYPE = "GTS_PDFA1"
|
|
OUTPUT_INTENT_REGISTRY = "http://www.color.org"
|
|
OUTPUT_INTENT_INFO = "sRGB IEC61966-2.1"
|
|
|
|
|
|
def _minimal_srgb_icc_profile() -> bytes:
|
|
"""
|
|
Build a minimal sRGB ICC profile that satisfies the PDF/A-3 requirement
|
|
for an embedded DestOutputProfile in the OutputIntent.
|
|
|
|
This is a stripped-down profile based on the sRGB IEC61966-2.1 spec.
|
|
It contains the required header, tag table, and enough data for veraPDF
|
|
to accept it as a valid ICC color profile.
|
|
"""
|
|
# ICC profile header (128 bytes)
|
|
header = bytearray(128)
|
|
# Profile size (filled in later)
|
|
# Preferred CMM type
|
|
header[4:8] = b"lcms"
|
|
# Profile version 2.1.0
|
|
header[8:12] = struct.pack(">I", 0x02100000)
|
|
# Device class: mntr (monitor)
|
|
header[12:16] = b"mntr"
|
|
# Color space: RGB
|
|
header[16:20] = b"RGB "
|
|
# PCS: XYZ
|
|
header[20:24] = b"XYZ "
|
|
# Date/time: 2024-01-01 00:00:00
|
|
header[24:36] = struct.pack(">6H", 2024, 1, 1, 0, 0, 0)
|
|
# Signature: acsp
|
|
header[36:40] = b"acsp"
|
|
# Primary platform: MSFT
|
|
header[40:44] = b"MSFT"
|
|
# Rendering intent: perceptual
|
|
header[64:68] = struct.pack(">I", 0)
|
|
# PCS illuminant D50 (X=0.9642, Y=1.0000, Z=0.8249 in s15Fixed16)
|
|
header[68:72] = struct.pack(">i", int(0.9642 * 65536))
|
|
header[72:76] = struct.pack(">i", int(1.0000 * 65536))
|
|
header[76:80] = struct.pack(">i", int(0.8249 * 65536))
|
|
# Creator signature
|
|
header[80:84] = b"lcms"
|
|
|
|
# Tag table: desc, wtpt, rXYZ, gXYZ, bXYZ, rTRC, gTRC, bTRC, cprt
|
|
tags = []
|
|
|
|
def _xyz_tag(x: float, y: float, z: float) -> bytes:
|
|
return (
|
|
b"XYZ "
|
|
+ b"\x00" * 4
|
|
+ struct.pack(">i", int(x * 65536))
|
|
+ struct.pack(">i", int(y * 65536))
|
|
+ struct.pack(">i", int(z * 65536))
|
|
)
|
|
|
|
def _curv_tag_gamma(gamma: float) -> bytes:
|
|
val = int(gamma * 256)
|
|
return b"curv" + b"\x00" * 4 + struct.pack(">I", 1) + struct.pack(">H", val) + b"\x00\x00"
|
|
|
|
def _desc_tag(text: str) -> bytes:
|
|
ascii_bytes = text.encode("ascii") + b"\x00"
|
|
data = b"desc" + b"\x00" * 4 + struct.pack(">I", len(ascii_bytes)) + ascii_bytes
|
|
# Unicode and ScriptCode localization (empty)
|
|
data += struct.pack(">I", 0) # Unicode language code
|
|
data += struct.pack(">I", 0) # Unicode count
|
|
data += struct.pack(">H", 0) + b"\x00" * 67 # ScriptCode
|
|
while len(data) % 4 != 0:
|
|
data += b"\x00"
|
|
return data
|
|
|
|
def _text_tag(text: str) -> bytes:
|
|
ascii_bytes = text.encode("ascii") + b"\x00"
|
|
data = b"text" + b"\x00" * 4 + ascii_bytes
|
|
while len(data) % 4 != 0:
|
|
data += b"\x00"
|
|
return data
|
|
|
|
# sRGB approximate values
|
|
desc_data = _desc_tag("sRGB IEC61966-2.1")
|
|
wtpt_data = _xyz_tag(0.9505, 1.0000, 1.0890)
|
|
rXYZ_data = _xyz_tag(0.4124, 0.2126, 0.0193)
|
|
gXYZ_data = _xyz_tag(0.3576, 0.7152, 0.1192)
|
|
bXYZ_data = _xyz_tag(0.1805, 0.0722, 0.9505)
|
|
rTRC_data = _curv_tag_gamma(2.2)
|
|
gTRC_data = _curv_tag_gamma(2.2)
|
|
bTRC_data = _curv_tag_gamma(2.2)
|
|
cprt_data = _text_tag("No copyright, use freely")
|
|
|
|
tag_datas = [
|
|
(b"desc", desc_data),
|
|
(b"wtpt", wtpt_data),
|
|
(b"rXYZ", rXYZ_data),
|
|
(b"gXYZ", gXYZ_data),
|
|
(b"bXYZ", bXYZ_data),
|
|
(b"rTRC", rTRC_data),
|
|
(b"gTRC", gTRC_data),
|
|
(b"bTRC", bTRC_data),
|
|
(b"cprt", cprt_data),
|
|
]
|
|
|
|
tag_count = len(tag_datas)
|
|
tag_table_size = 4 + tag_count * 12 # count + entries
|
|
data_offset = 128 + tag_table_size
|
|
|
|
tag_table = struct.pack(">I", tag_count)
|
|
payload = b""
|
|
for sig, data in tag_datas:
|
|
offset = data_offset + len(payload)
|
|
tag_table += sig + struct.pack(">II", offset, len(data))
|
|
payload += data
|
|
while len(payload) % 4 != 0:
|
|
payload += b"\x00"
|
|
|
|
profile = bytes(header) + tag_table + payload
|
|
# Write profile size into header
|
|
profile = struct.pack(">I", len(profile)) + profile[4:]
|
|
|
|
return profile
|
|
|
|
|
|
def _ensure_pdfa3_xmp(xmp_str: str) -> str:
|
|
"""Inject or update PDF/A-3 identification in XMP."""
|
|
pdfa_desc = (
|
|
f'<rdf:Description rdf:about="" xmlns:pdfaid="{PDFA_NS}">'
|
|
f"<pdfaid:part>{PDFA_PART}</pdfaid:part>"
|
|
f"<pdfaid:conformance>{PDFA_CONFORMANCE}</pdfaid:conformance>"
|
|
"</rdf:Description>"
|
|
)
|
|
if "pdfaid:part" in xmp_str and "pdfaid:conformance" in xmp_str:
|
|
return xmp_str
|
|
marker = "</rdf:RDF>"
|
|
if marker in xmp_str:
|
|
insert_pos = xmp_str.rfind(marker)
|
|
return xmp_str[:insert_pos] + pdfa_desc + "\n " + xmp_str[insert_pos:]
|
|
return xmp_str
|
|
|
|
|
|
def convert_to_pdfa3(pdf_bytes: bytes) -> Tuple[bytes, Optional[str]]:
|
|
"""
|
|
Normalize PDF to PDF/A-3b by adding identification XMP, an output intent
|
|
with an embedded sRGB ICC profile, and marking info as XMP-only.
|
|
|
|
Returns (new_pdf_bytes, None) on success, or (original_pdf_bytes, error_message) on failure.
|
|
"""
|
|
try:
|
|
import pikepdf
|
|
except ImportError as e:
|
|
return pdf_bytes, f"pikepdf not available: {e}"
|
|
|
|
try:
|
|
pdf = pikepdf.open(io.BytesIO(pdf_bytes))
|
|
except Exception as e:
|
|
return pdf_bytes, f"Invalid PDF: {e}"
|
|
|
|
try:
|
|
# Ensure metadata stream exists
|
|
if not hasattr(pdf.Root, "Metadata") or pdf.Root.Metadata is None:
|
|
minimal = (
|
|
'<?xpacket begin="\xef\xbb\xbf" id="W5M0MpCehiHzreSzNTczkc9d"?>'
|
|
'<x:xmpmeta xmlns:x="adobe:ns:meta/">'
|
|
'<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">'
|
|
"</rdf:RDF></x:xmpmeta>"
|
|
'<?xpacket end="w"?>'
|
|
)
|
|
pdf.Root.Metadata = pdf.make_stream(minimal.encode("utf-8"))
|
|
|
|
xmp_bytes = pdf.Root.Metadata.read_bytes()
|
|
xmp_str = xmp_bytes.decode("utf-8", errors="replace")
|
|
new_xmp = _ensure_pdfa3_xmp(xmp_str)
|
|
pdf.Root.Metadata = pdf.make_stream(new_xmp.encode("utf-8"))
|
|
|
|
# Add OutputIntent with embedded ICC profile for PDF/A-3
|
|
try:
|
|
intents = pdf.Root.get("/OutputIntents")
|
|
has_intent = intents is not None and len(intents) > 0
|
|
except Exception:
|
|
has_intent = False
|
|
|
|
if not has_intent:
|
|
try:
|
|
from pikepdf import Array, Dictionary, Name, Stream
|
|
|
|
icc_data = _minimal_srgb_icc_profile()
|
|
icc_stream = Stream(pdf, icc_data)
|
|
icc_stream[Name.N] = 3 # RGB = 3 components
|
|
|
|
intent = Dictionary(
|
|
Type=Name.OutputIntent,
|
|
S=Name("/GTS_PDFA1"),
|
|
OutputConditionIdentifier=OUTPUT_INTENT_INFO,
|
|
Info=OUTPUT_INTENT_INFO,
|
|
OutputCondition="sRGB IEC61966-2.1",
|
|
RegistryName=OUTPUT_INTENT_REGISTRY,
|
|
DestOutputProfile=icc_stream,
|
|
)
|
|
pdf.Root.OutputIntents = Array([intent])
|
|
except Exception:
|
|
# Fallback: intent without embedded profile (less compliant but still useful)
|
|
try:
|
|
from pikepdf import Array, Dictionary, Name
|
|
|
|
intent = Dictionary(
|
|
Type=Name.OutputIntent,
|
|
S=Name("/GTS_PDFA1"),
|
|
OutputConditionIdentifier=OUTPUT_INTENT_INFO,
|
|
Info=OUTPUT_INTENT_INFO,
|
|
OutputCondition="sRGB IEC61966-2.1",
|
|
RegistryName=OUTPUT_INTENT_REGISTRY,
|
|
)
|
|
pdf.Root.OutputIntents = Array([intent])
|
|
except Exception:
|
|
pass
|
|
|
|
out = io.BytesIO()
|
|
pdf_version = ("1", 7)
|
|
try:
|
|
pdf.save(
|
|
out,
|
|
min_version=pdf_version,
|
|
force_version=pdf_version,
|
|
fix_metadata_version=False,
|
|
)
|
|
except Exception as ex:
|
|
if "tuple" in str(ex).lower():
|
|
pdf.save(
|
|
out,
|
|
min_version=pdf_version,
|
|
force_version=None,
|
|
fix_metadata_version=False,
|
|
)
|
|
else:
|
|
raise
|
|
pdf.close()
|
|
return out.getvalue(), None
|
|
except Exception as e:
|
|
try:
|
|
pdf.close()
|
|
except Exception:
|
|
pass
|
|
return pdf_bytes, f"PDF/A-3 conversion failed: {e}"
|