From 6db5c39d78fd423bd4dc5ba9a09792b5dd9d5625 Mon Sep 17 00:00:00 2001
From: Eike Kettner <eike.kettner@posteo.de>
Date: Sun, 8 Nov 2020 01:23:07 +0100
Subject: [PATCH] Fix converted filename
Mark it by default with a string from the config file.
Issue: 397
---
.../src/main/scala/docspell/convert/ConvertConfig.scala | 1 +
.../src/test/scala/docspell/convert/ConversionTest.scala | 1 +
modules/joex/src/main/resources/reference.conf | 5 +++++
.../src/main/scala/docspell/joex/process/ConvertPdf.scala | 6 +++++-
4 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/modules/convert/src/main/scala/docspell/convert/ConvertConfig.scala b/modules/convert/src/main/scala/docspell/convert/ConvertConfig.scala
index f51791c0..8013f3bb 100644
--- a/modules/convert/src/main/scala/docspell/convert/ConvertConfig.scala
+++ b/modules/convert/src/main/scala/docspell/convert/ConvertConfig.scala
@@ -6,6 +6,7 @@ import docspell.convert.flexmark.MarkdownConfig
case class ConvertConfig(
chunkSize: Int,
+ convertedFilenamePart: String,
maxImageSize: Int,
markdown: MarkdownConfig,
wkhtmlpdf: WkHtmlPdfConfig,
diff --git a/modules/convert/src/test/scala/docspell/convert/ConversionTest.scala b/modules/convert/src/test/scala/docspell/convert/ConversionTest.scala
index 31dadd88..4d7e80ed 100644
--- a/modules/convert/src/test/scala/docspell/convert/ConversionTest.scala
+++ b/modules/convert/src/test/scala/docspell/convert/ConversionTest.scala
@@ -23,6 +23,7 @@ object ConversionTest extends SimpleTestSuite with FileChecks {
val convertConfig = ConvertConfig(
8192,
+ "converted",
3000 * 3000,
MarkdownConfig("body { padding: 2em 5em; }"),
WkHtmlPdfConfig(
diff --git a/modules/joex/src/main/resources/reference.conf b/modules/joex/src/main/resources/reference.conf
index 23ec5b47..51ad7d04 100644
--- a/modules/joex/src/main/resources/reference.conf
+++ b/modules/joex/src/main/resources/reference.conf
@@ -328,6 +328,11 @@ docspell.joex {
# as used with the rest server.
chunk-size = 524288
+ # A string used to change the filename of the converted pdf file.
+ # If empty, the original file name is used for the pdf file ( the
+ # extension is always replaced with `pdf`).
+ converted-filename-part = "converted"
+
# When reading images, this is the maximum size. Images that are
# larger are not processed.
max-image-size = ${docspell.joex.extraction.ocr.max-image-size}
diff --git a/modules/joex/src/main/scala/docspell/joex/process/ConvertPdf.scala b/modules/joex/src/main/scala/docspell/joex/process/ConvertPdf.scala
index 17cca3e0..65ff0dda 100644
--- a/modules/joex/src/main/scala/docspell/joex/process/ConvertPdf.scala
+++ b/modules/joex/src/main/scala/docspell/joex/process/ConvertPdf.scala
@@ -135,7 +135,11 @@ object ConvertPdf {
) = {
val hint =
MimeTypeHint.advertised(MimeType.pdf).withName(ra.name.getOrElse("file.pdf"))
- val newName = ra.name.map(n => s"$n.pdf")
+ val newName =
+ ra.name
+ .map(FileName.apply)
+ .map(_.withExtension("pdf").withPart(cfg.convertedFilenamePart, '.'))
+ .map(_.fullName)
ctx.store.bitpeace
.saveNew(pdf, cfg.chunkSize, MimetypeHint(hint.filename, hint.advertised))
.compile