From 7e1223c12c295152ddb245e8e51a870f1f5659f0 Mon Sep 17 00:00:00 2001
From: "James R. Barlow" <james@purplerock.ca>
Date: Sun, 29 Nov 2020 14:53:35 -0800
Subject: [PATCH] ghostscript: add output tracing

---
 src/ocrmypdf/_exec/ghostscript.py  | 43 +++++++++++++++++++++++++-----
 tests/plugins/gs_pdfa_failure.py   |  6 ++---
 tests/plugins/gs_render_failure.py |  2 +-
 3 files changed, 41 insertions(+), 10 deletions(-)

diff --git a/src/ocrmypdf/_exec/ghostscript.py b/src/ocrmypdf/_exec/ghostscript.py
index 0644e0d3..bbf13277 100644
--- a/src/ocrmypdf/_exec/ghostscript.py
+++ b/src/ocrmypdf/_exec/ghostscript.py
@@ -18,10 +18,11 @@ from subprocess import PIPE, CalledProcessError
 from typing import Optional, cast
 
 from PIL import Image
+from tqdm import tqdm
 
 from ocrmypdf.exceptions import MissingDependencyError, SubprocessOutputError
 from ocrmypdf.helpers import Resolution
-from ocrmypdf.subprocess import get_version, run
+from ocrmypdf.subprocess import get_version, run, run_polling_stderr
 
 log = logging.getLogger(__name__)
 
@@ -139,6 +140,27 @@ def rasterize_pdf(
         im.save(fspath(output_file), dpi=page_dpi)
 
 
+class GhostscriptFollower:
+    re_process = re.compile(r"Processing pages \d+ through (\d+).")
+    re_page = re.compile(r"Page (\d+)")
+
+    def __init__(self):
+        self.count = 0
+        self.tqdm = None
+
+    def __call__(self, line):
+        if not self.tqdm:
+            m = self.re_process.match(line.strip())
+            if m:
+                self.count = int(m.group(1))
+                self.tqdm = tqdm(total=self.count, desc="Ghostscript", unit='page')
+                return
+        else:
+            m = self.re_page.match(line.strip())
+            if m:
+                self.tqdm.update()
+
+
 def generate_pdfa(
     pdf_pages,
     output_file: os.PathLike,
@@ -188,7 +210,6 @@ def generate_pdfa(
     args_gs = (
         [
             GS,
-            "-dQUIET",
             "-dBATCH",
             "-dNOPAUSE",
             "-dSAFER",
@@ -208,16 +229,26 @@ def generate_pdfa(
         ]
     )
     args_gs.extend(fspath(s) for s in pdf_pages)  # Stringify Path objs
+
     try:
         with Path(output_file).open('wb') as output:
-            p = run(args_gs, stdout=output, stderr=PIPE, check=True)
+            p = run_polling_stderr(
+                args_gs,
+                stdout=output,
+                stderr=PIPE,
+                check=True,
+                text=True,
+                encoding='utf-8',
+                errors='replace',
+                callback=GhostscriptFollower(),
+            )
     except CalledProcessError as e:
         # Ghostscript does not change return code when it fails to create
         # PDF/A - check PDF/A status elsewhere
-        log.error(e.stderr.decode(errors='replace'))
-        raise SubprocessOutputError('Ghostscript PDF/A rendering failed')
+        log.error(e.stderr)
+        raise SubprocessOutputError('Ghostscript PDF/A rendering failed') from e
     else:
-        stderr = p.stderr.decode('utf-8', errors='replace')
+        stderr = p.stderr
         if _gs_error_reported(stderr):
             last_part = None
             repcount = 0
diff --git a/tests/plugins/gs_pdfa_failure.py b/tests/plugins/gs_pdfa_failure.py
index dcad94f6..8a694de9 100644
--- a/tests/plugins/gs_pdfa_failure.py
+++ b/tests/plugins/gs_pdfa_failure.py
@@ -23,7 +23,7 @@ from unittest.mock import patch
 
 from ocrmypdf import hookimpl
 from ocrmypdf.builtin_plugins import ghostscript
-from ocrmypdf.subprocess import run
+from ocrmypdf.subprocess import run_polling_stderr
 
 
 def run_rig_args(args, **kwargs):
@@ -33,13 +33,13 @@ def run_rig_args(args, **kwargs):
     new_args = [
         arg for arg in args if not arg.startswith('-dPDFA') and not arg.endswith('.ps')
     ]
-    proc = run(new_args, **kwargs)
+    proc = run_polling_stderr(new_args, **kwargs)
     return proc
 
 
 @hookimpl
 def generate_pdfa(pdf_pages, pdfmark, output_file, compression, pdf_version, pdfa_part):
-    with patch('ocrmypdf._exec.ghostscript.run', new=run_rig_args):
+    with patch('ocrmypdf._exec.ghostscript.run_polling_stderr', new=run_rig_args):
         ghostscript.generate_pdfa(
             pdf_pages=pdf_pages,
             pdfmark=pdfmark,
diff --git a/tests/plugins/gs_render_failure.py b/tests/plugins/gs_render_failure.py
index e3cee162..2cad1f4e 100644
--- a/tests/plugins/gs_render_failure.py
+++ b/tests/plugins/gs_render_failure.py
@@ -34,7 +34,7 @@ def raise_gs_fail(*args, **kwargs):
 
 @hookimpl
 def generate_pdfa(pdf_pages, pdfmark, output_file, compression, pdf_version, pdfa_part):
-    with patch('ocrmypdf._exec.ghostscript.run', new=raise_gs_fail):
+    with patch('ocrmypdf._exec.ghostscript.run_polling_stderr', new=raise_gs_fail):
         ghostscript.generate_pdfa(
             pdf_pages=pdf_pages,
             pdfmark=pdfmark,