summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorNick White <git@njw.name>2019-12-11 15:47:35 +0000
committerNick White <git@njw.name>2019-12-11 15:47:35 +0000
commitaee651e282c7a9aecd8f25a5e306cdd695325068 (patch)
tree8f12712ef74365185914f29ac557f1bf09005bc8
parent23c1b47eb42ef8de6f5d5dd8e4afb80e8bd17497 (diff)
Clean up and correct book name parsing in the pipeline, and update usage of getpipelinebook
-rw-r--r--cmd/bookpipeline/main.go17
-rw-r--r--cmd/getpipelinebook/main.go6
2 files changed, 6 insertions, 17 deletions
diff --git a/cmd/bookpipeline/main.go b/cmd/bookpipeline/main.go
index 0ffb4b2..425b6a6 100644
--- a/cmd/bookpipeline/main.go
+++ b/cmd/bookpipeline/main.go
@@ -454,13 +454,7 @@ func ocrPage(msg bookpipeline.Qmsg, conn Pipeliner, process func(chan string, ch
errc := make(chan error)
msgparts := strings.Split(msg.Body, " ")
- bookparts := strings.Split(msgparts[0], "/")
- var bookname string
- if len(bookparts) > 1 {
- bookname = filepath.Dir(msgparts[0])
- } else {
- bookname = msgparts[0]
- }
+ bookname := filepath.Dir(msgparts[0])
if len(msgparts) > 1 {
process = ocr(msgparts[1])
}
@@ -538,13 +532,8 @@ func processBook(msg bookpipeline.Qmsg, conn Pipeliner, process func(chan string
errc := make(chan error)
msgparts := strings.Split(msg.Body, " ")
- bookparts := strings.Split(msgparts[0], "/")
- var bookname string
- if len(bookparts) > 1 {
- bookname = filepath.Dir(msgparts[0])
- } else {
- bookname = msgparts[0]
- }
+ bookname := msgparts[0]
+
var training string
if len(msgparts) > 1 {
training = msgparts[1]
diff --git a/cmd/getpipelinebook/main.go b/cmd/getpipelinebook/main.go
index faf5383..fc6170c 100644
--- a/cmd/getpipelinebook/main.go
+++ b/cmd/getpipelinebook/main.go
@@ -15,9 +15,9 @@ const usage = `Usage: getpipelinebook [-a] [-c] [-v] bookname
Downloads the pipeline results for a book.
-By default this downloads the best hOCR version for each page with
-the corresponding binarised image, and the best, conf and graph.png
-analysis files.
+By default this downloads the best hOCR version for each page, the
+binarised and (if available) colour PDF, and the best, conf and
+graph.png analysis files.
`
// null writer to enable non-verbose logging to be discarded