WIP for progress report

dangerzone/ctx.py

@@ -0,0 +1,137 @@
+import datetime
+import enum
+import logging
+import time
+from typing import Callable
+
+from colorama import Fore, Style
+
+from .document import Document
+
+log = logging.getLogger(__name__)
+
+
+class ConversionCtx:
+
+    EST_PERCENT_START_CONVERSION_PROC = 1
+    EST_PERCENT_GATHER_PAGES = 2
+    EST_PERCENT_CONVERT_PAGES = 96
+    EST_PERCENT_COMPLETE_CONVERSION = 1
+
+    MSG_CONVERSION_PROCESS_TYPE = "process"
+
+    # Conversion state
+    STATE_NOT_STARTED = enum.auto()
+    STATE_STARTING_CONVERSION_PROC = enum.auto()
+    STATE_GATHERING_PAGES = enum.auto()
+    STATE_CONVERTING_PAGES = enum.auto()
+    STATE_COMPLETED = enum.auto()
+    STATE_FAILED = enum.auto()
+
+    def __init__(
+        self,
+        document: Document,
+        ocr_lang: str | None = None,
+        progress_callback: Callable | None = None,
+    ) -> None:
+        self.doc = document
+        self.ocr_lang = ocr_lang
+        self.callback = progress_callback
+
+        conversion_total = 100  # FiXME:
+        assert conversion_total == 100
+
+        self.percentage: float = 0.0
+        self.cur_page = 0
+        self.pages = 0
+        self.page_timer_start = None
+        self.state = self.STATE_NOT_STARTED
+
+    def is_not_started(self) -> bool:
+        return self.state is self.STATE_NOT_STARTED
+
+    def is_started(self) -> bool:
+        return self.state in (
+            self.STATE_STARTING_CONVERSION_PROC,
+            self.STATE_GATHERING_PAGES,
+            self.STATE_CONVERTING_PAGES,
+        )
+
+    def is_completed(self) -> bool:
+        return self.state is Document.STATE_COMPLETED
+
+    def is_failed(self) -> bool:
+        return self.state is Document.STATE_FAILED
+
+    def increase(self, step: float) -> None:
+        assert step > 0
+        self.percentage += step
+
+    def print_message(self, text: str, error: bool = False) -> None:
+        s = Style.BRIGHT + Fore.YELLOW + f"[doc {self.doc.id}] "
+        s += Fore.CYAN + f"{int(self.percentage)}% " + Style.RESET_ALL
+        if error:
+            s += Fore.RED + text + Style.RESET_ALL
+            log.error(s)
+        else:
+            s += text
+            log.info(s)
+
+        if self.callback:
+            self.callback(error, text, self.percentage)
+
+    def start_conversion_proc(self):
+        self.state = self.STATE_STARTING_CONVERSION_PROC
+        self.print_message(
+            f"Starting a {self.MSG_CONVERSION_PROCESS_TYPE} for the document conversion"
+        )
+
+    def start_page_gathering(self):
+        self.state = self.STATE_GATHERING_PAGES
+        self.increase(self.EST_PERCENT_START_CONVERSION_PROC)
+        self.print_message("Getting number of pages")
+
+    def set_total_pages(self, pages: int) -> None:
+        self.state = self.STATE_CONVERTING_PAGES
+        self.increase(self.EST_PERCENT_GATHER_PAGES)
+        assert pages > 0
+        self.pages = pages
+
+    def page_iter(self, pages):
+        self.set_total_pages(pages)
+        for page in range(1, pages + 1):
+            self.start_converting_page(page)
+            yield page
+            self.finished_converting_page()
+
+    def start_converting_page(self, page: int) -> None:
+        searchable = "searchable " if self.ocr_lang else ""
+        remaining = ""
+
+        if not self.page_timer_start:
+            self.page_timer_start = time.monotonic()
+        else:
+            processed_pages = page - 1
+            elapsed = time.monotonic() - self.page_timer_start
+            elapsed_per_page = elapsed / processed_pages
+            remaining = (self.pages - processed_pages) * elapsed_per_page
+            remaining = datetime.timedelta(seconds=round(remaining))
+            remaining = f" (remaining: {remaining}s)"
+
+        self.print_message(
+            f"Converting page {page}/{self.pages} from pixels to {searchable}PDF{remaining}"
+        )
+
+    def finished_converting_page(self) -> None:
+        self.increase(self.EST_PERCENT_CONVERT_PAGES / self.pages)
+
+    def fail(self, msg: str) -> None:
+        self.state = self.STATE_FAILED
+        self.print_message(msg, error=True)
+        self.doc.mark_as_failed()
+
+    def success(self) -> None:
+        self.state = self.STATE_COMPLETED
+        self.percentage = 100
+        self.doc.mark_as_safe()
+        self.print_message("Conversion completed successfully")

dangerzone/gui/main_window.py

@@ -29,6 +29,7 @@ from ..isolation_provider.container import Container, NoContainerTechException
 from ..isolation_provider.dummy import Dummy
 from ..isolation_provider.qubes import Qubes, is_qubes_native_conversion
 from ..util import get_resource_path, get_subprocess_startupinfo, get_version
+from ..ctx import ConversionCtx
 from .logic import Alert, CollapsibleBox, DangerzoneGui, UpdateDialog
 from .updater import UpdateReport
 
@@ -1124,11 +1125,8 @@ class ConvertTask(QtCore.QObject):
         self.dangerzone = dangerzone
 
     def convert_document(self) -> None:
-        self.dangerzone.isolation_provider.convert(
+        ctx = ConversionCtx(self.document, self.ocr_lang, self.progress_callback)
-            self.document,
+        self.dangerzone.isolation_provider.convert(ctx)
-            self.ocr_lang,
-            self.progress_callback,
-        )
         self.finished.emit(self.error)
 
     def progress_callback(self, error: bool, text: str, percentage: int) -> None:

dangerzone/isolation_provider/base.py

@@ -16,6 +16,7 @@ from ..conversion import errors
 from ..conversion.common import DEFAULT_DPI, INT_BYTES
 from ..document import Document
 from ..util import get_tessdata_dir, replace_control_chars
+from ..ctx import ConversionCtx
 
 log = logging.getLogger(__name__)
 
@@ -97,29 +98,24 @@ class IsolationProvider(ABC):
     def install(self) -> bool:
         pass
 
-    def convert(
+    def convert(self, ctx: ConversionCtx) -> None:
-        self,
+        document = ctx.doc
-        document: Document,
-        ocr_lang: Optional[str],
-        progress_callback: Optional[Callable] = None,
-    ) -> None:
-        self.progress_callback = progress_callback
         document.mark_as_converting()
         try:
+            ctx.start_conversion_proc()
             with self.doc_to_pixels_proc(document) as conversion_proc:
-                self.convert_with_proc(document, ocr_lang, conversion_proc)
+                ctx.start_page_gathering()
-            document.mark_as_safe()
+                self.convert_with_proc(ctx, conversion_proc)
             if document.archive_after_conversion:
                 document.archive()
+            ctx.success()
         except errors.ConversionException as e:
-            self.print_progress(document, True, str(e), 0)
+            ctx.fail(str(e))
-            document.mark_as_failed()
         except Exception as e:
             log.exception(
                 f"An exception occurred while converting document '{document.id}'"
             )
-            self.print_progress(document, True, str(e), 0)
+            ctx.fail(str(e))
-            document.mark_as_failed()
 
     def ocr_page(self, pixmap: fitz.Pixmap, ocr_lang: str) -> bytes:
         """Get a single page as pixels, OCR it, and return a PDF as bytes."""
@@ -157,12 +153,13 @@ class IsolationProvider(ABC):
 
     def convert_with_proc(
         self,
-        document: Document,
+        ctx: ConversionCtx,
-        ocr_lang: Optional[str],
         p: subprocess.Popen,
     ) -> None:
+        ocr_lang = ctx.ocr_lang
+        document = ctx.doc
         percentage = 0.0
-        with open(document.input_filename, "rb") as f:
+        with open(ctx.doc.input_filename, "rb") as f:
             try:
                 assert p.stdin is not None
                 p.stdin.write(f.read())
@@ -178,13 +175,7 @@ class IsolationProvider(ABC):
 
             safe_doc = fitz.Document()
 
-            for page in range(1, n_pages + 1):
+            for page in ctx.page_iter(n_pages):
-                searchable = "searchable " if ocr_lang else ""
-                text = (
-                    f"Converting page {page}/{n_pages} from pixels to {searchable}PDF"
-                )
-                self.print_progress(document, False, text, percentage)
-
                 width = read_int(p.stdout)
                 height = read_int(p.stdout)
                 if not (1 <= width <= errors.MAX_PAGE_WIDTH):
@@ -216,25 +207,6 @@ class IsolationProvider(ABC):
         safe_doc.save(document.sanitized_output_filename)
         os.replace(document.sanitized_output_filename, document.output_filename)
 
-        # TODO handle leftover code input
-        text = "Successfully converted document"
-        self.print_progress(document, False, text, 100)
-
-    def print_progress(
-        self, document: Document, error: bool, text: str, percentage: float
-    ) -> None:
-        s = Style.BRIGHT + Fore.YELLOW + f"[doc {document.id}] "
-        s += Fore.CYAN + f"{int(percentage)}% " + Style.RESET_ALL
-        if error:
-            s += Fore.RED + text + Style.RESET_ALL
-            log.error(s)
-        else:
-            s += text
-            log.info(s)
-
-        if self.progress_callback:
-            self.progress_callback(error, text, percentage)
-
     def get_proc_exception(
         self, p: subprocess.Popen, timeout: int = TIMEOUT_EXCEPTION
     ) -> Exception:

dangerzone/logic.py

@@ -7,6 +7,7 @@ import colorama
 
 from . import errors, util
 from .document import Document
+from .ctx import ConversionCtx
 from .isolation_provider.base import IsolationProvider
 from .settings import Settings
 from .util import get_resource_path
@@ -65,12 +66,9 @@ class DangerzoneCore(object):
         self, ocr_lang: Optional[str], stdout_callback: Optional[Callable] = None
     ) -> None:
         def convert_doc(document: Document) -> None:
+            ctx = ConversionCtx(document, ocr_lang, stdout_callback)
             try:
-                self.isolation_provider.convert(
+                self.isolation_provider.convert(ctx)
-                    document,
-                    ocr_lang,
-                    stdout_callback,
-                )
             except Exception as e:
                 log.exception(
                     f"Unexpected error occurred while converting '{document}'"

prog_tests.py

@@ -0,0 +1,26 @@
+#!/usr/bin/python3
+
+import logging
+import sys
+import time
+
+from dangerzone import document
+from dangerzone.ctx import ConversionCtx
+
+
+def main():
+    logging.basicConfig(level=logging.INFO)
+    doc = document.Document()
+    ctx = ConversionCtx(doc)
+    ctx.start_conversion_proc()
+    ctx.start_page_gathering()
+    for page in ctx.page_iter(10):
+        time.sleep(0.2)
+        if not page % 5:
+            ctx.fail(f"Failed during page {page}")
+
+    ctx.success()
+
+
+if __name__ == "__main__":
+    sys.exit(main())