mirror of
https://github.com/paperless-ngx/paperless-ngx.git
synced 2026-10-20 23:08:14 +00:00
280 lines
11 KiB
Python
280 lines
11 KiB
Python
# Generated by Django 5.2.14 on 2026-06-16 17:36
|
|
|
|
import django.core.validators
|
|
import django.db.models.deletion
|
|
import django.utils.timezone
|
|
from django.conf import settings
|
|
from django.db import migrations
|
|
from django.db import models
|
|
|
|
|
|
class Migration(migrations.Migration):
|
|
dependencies = [
|
|
("documents", "0027_documentbarcode"),
|
|
]
|
|
|
|
operations = [
|
|
migrations.CreateModel(
|
|
name="OcrTemplate",
|
|
fields=[
|
|
(
|
|
"id",
|
|
models.AutoField(
|
|
auto_created=True,
|
|
primary_key=True,
|
|
serialize=False,
|
|
verbose_name="ID",
|
|
),
|
|
),
|
|
("name", models.CharField(max_length=128, verbose_name="name")),
|
|
(
|
|
"owner",
|
|
models.ForeignKey(
|
|
blank=True,
|
|
default=None,
|
|
null=True,
|
|
on_delete=django.db.models.deletion.SET_NULL,
|
|
to=settings.AUTH_USER_MODEL,
|
|
verbose_name="owner",
|
|
),
|
|
),
|
|
(
|
|
"source_width",
|
|
models.PositiveIntegerField(
|
|
help_text="Width of the image the zones were drawn on (px)",
|
|
validators=[django.core.validators.MinValueValidator(1)],
|
|
verbose_name="source width",
|
|
),
|
|
),
|
|
(
|
|
"source_height",
|
|
models.PositiveIntegerField(
|
|
help_text="Height of the image the zones were drawn on (px)",
|
|
validators=[django.core.validators.MinValueValidator(1)],
|
|
verbose_name="source height",
|
|
),
|
|
),
|
|
("enabled", models.BooleanField(default=True, verbose_name="enabled")),
|
|
(
|
|
"combine_formats",
|
|
models.JSONField(
|
|
blank=True,
|
|
default=dict,
|
|
help_text="Per-target format strings for combining several zones into one field, keyed by target (custom field id, or 'title'/'asn'/'created'). Tokens like {Zone Name} are replaced with that zone's value.",
|
|
verbose_name="combine formats",
|
|
),
|
|
),
|
|
(
|
|
"created",
|
|
models.DateTimeField(
|
|
db_index=True,
|
|
default=django.utils.timezone.now,
|
|
editable=False,
|
|
verbose_name="created",
|
|
),
|
|
),
|
|
(
|
|
"updated",
|
|
models.DateTimeField(auto_now=True, verbose_name="updated"),
|
|
),
|
|
(
|
|
"document_type",
|
|
models.ForeignKey(
|
|
on_delete=django.db.models.deletion.CASCADE,
|
|
related_name="ocr_templates",
|
|
to="documents.documenttype",
|
|
verbose_name="document type",
|
|
),
|
|
),
|
|
(
|
|
"sample_document",
|
|
models.ForeignKey(
|
|
blank=True,
|
|
help_text="Document used for previewing zones in the editor",
|
|
null=True,
|
|
on_delete=django.db.models.deletion.SET_NULL,
|
|
related_name="+",
|
|
to="documents.document",
|
|
verbose_name="sample document",
|
|
),
|
|
),
|
|
],
|
|
options={
|
|
"verbose_name": "OCR template",
|
|
"verbose_name_plural": "OCR templates",
|
|
"ordering": ("name",),
|
|
},
|
|
),
|
|
migrations.CreateModel(
|
|
name="OcrTemplateZone",
|
|
fields=[
|
|
(
|
|
"id",
|
|
models.AutoField(
|
|
auto_created=True,
|
|
primary_key=True,
|
|
serialize=False,
|
|
verbose_name="ID",
|
|
),
|
|
),
|
|
(
|
|
"name",
|
|
models.CharField(
|
|
help_text="Descriptive name for this zone (e.g. 'Invoice Number')",
|
|
max_length=128,
|
|
verbose_name="zone name",
|
|
),
|
|
),
|
|
(
|
|
"target",
|
|
models.CharField(
|
|
choices=[
|
|
("custom_field", "Custom field"),
|
|
("title", "Title"),
|
|
("asn", "Archive serial number"),
|
|
("created", "Date created"),
|
|
],
|
|
default="custom_field",
|
|
help_text="Where the extracted value is written: a custom field, or a built-in document field (title, ASN, created date)",
|
|
max_length=20,
|
|
verbose_name="target",
|
|
),
|
|
),
|
|
(
|
|
"page",
|
|
models.IntegerField(
|
|
blank=True,
|
|
help_text="Page (1 = first, -1 = last; blank uses the template default)",
|
|
null=True,
|
|
verbose_name="page",
|
|
),
|
|
),
|
|
(
|
|
"x",
|
|
models.PositiveIntegerField(
|
|
help_text="Left edge (px)",
|
|
verbose_name="x",
|
|
),
|
|
),
|
|
(
|
|
"y",
|
|
models.PositiveIntegerField(
|
|
help_text="Top edge (px)",
|
|
verbose_name="y",
|
|
),
|
|
),
|
|
(
|
|
"width",
|
|
models.PositiveIntegerField(
|
|
help_text="Zone width (px)",
|
|
validators=[django.core.validators.MinValueValidator(1)],
|
|
verbose_name="width",
|
|
),
|
|
),
|
|
(
|
|
"height",
|
|
models.PositiveIntegerField(
|
|
help_text="Zone height (px)",
|
|
validators=[django.core.validators.MinValueValidator(1)],
|
|
verbose_name="height",
|
|
),
|
|
),
|
|
(
|
|
"zone_source_width",
|
|
models.PositiveIntegerField(
|
|
blank=True,
|
|
help_text="Width of the page image this zone was drawn on (px). Falls back to template source_width if unset.",
|
|
null=True,
|
|
verbose_name="zone source width",
|
|
),
|
|
),
|
|
(
|
|
"zone_source_height",
|
|
models.PositiveIntegerField(
|
|
blank=True,
|
|
help_text="Height of the page image this zone was drawn on (px). Falls back to template source_height if unset.",
|
|
null=True,
|
|
verbose_name="zone source height",
|
|
),
|
|
),
|
|
(
|
|
"ocr_language",
|
|
models.CharField(
|
|
default="eng",
|
|
help_text="Tesseract language code(s), e.g. 'deu+eng'",
|
|
max_length=20,
|
|
verbose_name="OCR language",
|
|
),
|
|
),
|
|
(
|
|
"transform",
|
|
models.CharField(
|
|
choices=[
|
|
("none", "None"),
|
|
("strip", "Strip whitespace"),
|
|
("uppercase", "Uppercase"),
|
|
("lowercase", "Lowercase"),
|
|
("numeric", "Numeric only"),
|
|
(
|
|
"strip_punctuation",
|
|
"Remove leading/trailing punctuation",
|
|
),
|
|
("date", "Parse date"),
|
|
("qr_code", "Read QR/barcode"),
|
|
],
|
|
default="strip",
|
|
max_length=20,
|
|
verbose_name="transform",
|
|
),
|
|
),
|
|
(
|
|
"date_format",
|
|
models.CharField(
|
|
blank=True,
|
|
default="",
|
|
help_text="Python strptime format for the 'Parse date' transform (e.g. %d.%m.%Y). Blank = auto-detect.",
|
|
max_length=64,
|
|
verbose_name="date format",
|
|
),
|
|
),
|
|
(
|
|
"validation_regex",
|
|
models.CharField(
|
|
blank=True,
|
|
default="",
|
|
help_text="Optional regex pattern — extracted text is only accepted if it matches",
|
|
max_length=256,
|
|
verbose_name="validation regex",
|
|
),
|
|
),
|
|
("order", models.PositiveIntegerField(default=0, verbose_name="order")),
|
|
(
|
|
"custom_field",
|
|
models.ForeignKey(
|
|
blank=True,
|
|
help_text="Target custom field (only used when target is 'custom_field')",
|
|
null=True,
|
|
on_delete=django.db.models.deletion.CASCADE,
|
|
related_name="ocr_zones",
|
|
to="documents.customfield",
|
|
verbose_name="custom field",
|
|
),
|
|
),
|
|
(
|
|
"template",
|
|
models.ForeignKey(
|
|
on_delete=django.db.models.deletion.CASCADE,
|
|
related_name="zones",
|
|
to="documents.ocrtemplate",
|
|
verbose_name="template",
|
|
),
|
|
),
|
|
],
|
|
options={
|
|
"verbose_name": "OCR template zone",
|
|
"verbose_name_plural": "OCR template zones",
|
|
"ordering": ("template", "order"),
|
|
},
|
|
),
|
|
]
|