Skip to contentSkip to conversation

First pass of cleanup of messages outputted to the user.

PATCH In review Chris Larson 4 participants 9 messages Apr 8, 2009 – Apr 9, 2009

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 First pass of cleanup of messages outputted to the user.

OpenEmbedded outputs a lot of messages that the user is likely to never care about. We should only output something when it reflects upon their recipe (i.e. unpacking their sources, applying their patches), or is quite significant or unusual.

---
 classes/base.bbclass             |   17 +++--------------
 classes/insane.bbclass           |   14 +++++++-------
 classes/package.bbclass          |    4 ++--
 classes/packaged-staging.bbclass |    6 +++---
 4 files changed, 15 insertions(+), 26 deletions(-)
classes/base.bbclass +3 -14
index 9ec705b..7cfebf2 100644
@@ -890,22 +890,11 @@ python base_eventhandler() {
 	from bb.event import Handled, NotHandled, getName
 	import os
 
-	messages = {}
-	messages["Completed"] = "completed"
-	messages["Succeeded"] = "completed"
-	messages["Started"] = "started"
-	messages["Failed"] = "failed"
-
 	name = getName(e)
-	msg = ""
-	if name.startswith("Task"):
-		msg += "package %s: task %s: " % (data.getVar("PF", e.data, 1), e.task)
-		msg += messages.get(name[4:]) or name[4:]
-	elif name.startswith("Build"):
-		msg += "build %s: " % e.name
-		msg += messages.get(name[5:]) or name[5:]
+	if name == "TaskCompleted":
+		msg = "package %s: task %s is complete." % (data.getVar("PF", e.data, 1), e.task)
 	elif name == "UnsatisfiedDep":
-		msg += "package %s: dependency %s %s" % (e.pkg, e.dep, name[:-3].lower())
+		msg = "package %s: dependency %s %s" % (e.pkg, e.dep, name[:-3].lower())
 	else:
 		return NotHandled
 
classes/insane.bbclass +7 -7
index 48964af..4496e15 100644
@@ -215,7 +215,7 @@ def package_qa_write_error(error_class, name, path, d):
 
 def package_qa_handle_error(error_class, error_msg, name, path, d):
     import bb
-    bb.error("QA Issue: %s" % error_msg)
+    bb.error("QA Issue with %s: %s" % (name, error_msg))
     package_qa_write_error(error_class, name, path, d)
     return not package_qa_make_fatal_error(error_class, name, path, d)
 
@@ -481,7 +481,7 @@ def package_qa_check_rdepends(pkg, workdir, d):
 # The PACKAGE FUNC to scan each package
 python do_package_qa () {
     import bb
-    bb.note("DO PACKAGE QA")
+    bb.debug(2, "DO PACKAGE QA")
     workdir = bb.data.getVar('WORKDIR', d, True)
     packages = bb.data.getVar('PACKAGES',d, True)
 
@@ -496,10 +496,10 @@ python do_package_qa () {
     rdepends_sane = True
     for package in packages.split():
         if bb.data.getVar('INSANE_SKIP_' + package, d, True):
-            bb.note("Package: %s (skipped)" % package)
+            bb.note("package %s skipped" % package)
             continue
 
-        bb.note("Checking Package: %s" % package)
+        bb.debug(1, "Checking Package: %s" % package)
         path = "%s/install/%s" % (workdir, package)
         if not package_qa_walk(path, checks, package, d):
             walk_sane  = False
@@ -508,14 +508,14 @@ python do_package_qa () {
 
     if not walk_sane or not rdepends_sane:
         bb.fatal("QA run found fatal errors. Please consider fixing them.")
-    bb.note("DONE with PACKAGE QA")
+    bb.debug(2, "DONE with PACKAGE QA")
 }
 
 
 # The Staging Func, to check all staging
 addtask qa_staging after do_populate_staging before do_build
 python do_qa_staging() {
-    bb.note("QA checking staging")
+    bb.debug(2, "QA checking staging")
 
     if not package_qa_check_staged(bb.data.getVar('STAGING_LIBDIR',d,True), d):
         bb.fatal("QA staging was broken by the package built above")
@@ -524,7 +524,7 @@ python do_qa_staging() {
 # Check broken config.log files
 addtask qa_configure after do_configure before do_compile
 python do_qa_configure() {
-    bb.note("Checking sanity of the config.log file")
+    bb.debug(1, "Checking sanity of the config.log file")
     import os
     for root, dirs, files in os.walk(bb.data.getVar('WORKDIR', d, True)):
         statement = "grep 'CROSS COMPILE Badness:' %s > /dev/null" % \
classes/package.bbclass +2 -2
index 7a61c5a..3a30d38 100644
@@ -166,7 +166,7 @@ def runstrip(file, d):
     # If the file is in a .debug directory it was already stripped,
     # don't do it again...
     if os.path.dirname(file).endswith(".debug"):
-        bb.note("Already ran strip")
+        bb.debug(2, "Already ran strip on %s" % file)
         return 0
 
     strip = bb.data.getVar("STRIP", d, 1)
@@ -560,7 +560,7 @@ python package_do_shlibs() {
 
 	exclude_shlibs = bb.data.getVar('EXCLUDE_FROM_SHLIBS', d, 0)
 	if exclude_shlibs:
-		bb.note("not generating shlibs")
+		bb.debug(1, "not generating shlibs")
 		return
 		
 	lib_re = re.compile("^lib.*\.so")
classes/packaged-staging.bbclass +3 -3
index 849f60c..82c1a0d 100644
@@ -109,17 +109,17 @@ def pstage_cleanpackage(pkgname, d):
 	pstage_set_pkgmanager(d)
 	list_cmd = bb.data.getVar("PSTAGE_LIST_CMD", d, True)
 
-	bb.note("Checking if staging package installed")
+	bb.debug(2, "Checking if staging package installed")
 	lf = bb.utils.lockfile(bb.data.expand("${STAGING_DIR}/staging.lock", d))
 	ret = os.system("PATH=\"%s\" %s | grep %s" % (path, list_cmd, pkgname))
 	if ret == 0:
-		bb.note("Yes. Uninstalling package from staging...")
+		bb.debug(1, "Uninstalling package from staging...")
 		removecmd = bb.data.getVar("PSTAGE_REMOVE_CMD", d, 1)
 		ret = os.system("PATH=\"%s\" %s %s" % (path, removecmd, pkgname))
 		if ret != 0:
 			bb.note("Failure removing staging package")
 	else:
-		bb.note("No. Manually removing any installed files")
+		bb.debug(1, "Manually removing any installed files from staging...")
 		pstage_manualclean("staging", "STAGING_DIR", d)
 		pstage_manualclean("cross", "CROSS_DIR", d)
 		pstage_manualclean("deploy", "DEPLOY_DIR", d)
Signature · 1 line
1.6.0.2
#1 lore raw

PATCH 1/1 Shorten some full paths printed to the user.

Adds a base_path_out convenience function, which prepares a full path for display to the user. The initial implementation just makes it relative to ${TOPDIR}. This function is then used for some messages outputted to the user (packaged-staging, patch application, clean, unpack tasks).

---
 classes/base.bbclass             |   17 +++++++++++++----
 classes/packaged-staging.bbclass |    2 +-
 classes/patch.bbclass            |    2 +-
 3 files changed, 15 insertions(+), 6 deletions(-)
classes/base.bbclass +13 -4
index 7cfebf2..500e39a 100644
@@ -39,6 +39,14 @@ def base_path_relative(src, dest):
 
     return sep.join(relpath)
 
+def base_path_out(path, d):
+    """ Prepare a path for display to the user. """
+    rel = base_path_relative(d.getVar("TOPDIR", 1), path)
+    if len(rel) > len(path):
+        return path
+    else:
+        return rel
+
 # for MD5/SHA handling
 def base_chk_load_parser(config_path):
     import ConfigParser, os, bb
@@ -70,6 +78,7 @@ def base_chk_file(parser, pn, pv, src_uri, localpath, data):
 
     # md5 and sha256 should be valid now
     if not os.path.exists(localpath):
+        localpath = base_path_out(localpath, data)
         bb.note("The localpath does not exist '%s'" % localpath)
         raise Exception("The path does not exist '%s'" % localpath)
 
@@ -496,11 +505,11 @@ python base_do_clean() {
 	"""clear the build and temp directories"""
 	dir = bb.data.expand("${WORKDIR}", d)
 	if dir == '//': raise bb.build.FuncFailed("wrong DATADIR")
-	bb.note("removing " + dir)
+	bb.note("removing " + base_path_out(dir, d))
 	os.system('rm -rf ' + dir)
 
 	dir = "%s.*" % bb.data.expand(bb.data.getVar('STAMP', d), d)
-	bb.note("removing " + dir)
+	bb.note("removing " + base_path_out(dir, d))
 	os.system('rm -f '+ dir)
 }
 
@@ -555,7 +564,7 @@ python base_do_distclean() {
 		except bb.MalformedUrl, e:
 			bb.debug(1, 'Unable to generate local path for malformed uri: %s' % e)
 		else:
-			bb.note("removing %s" % local)
+			bb.note("removing %s" % base_path_out(local, d))
 			try:
 				if os.path.exists(local + ".md5"):
 					os.remove(local + ".md5")
@@ -771,7 +780,7 @@ def oe_unpack_file(file, data, url = None):
 		os.chdir(newdir)
 
 	cmd = "PATH=\"%s\" %s" % (bb.data.getVar('PATH', data, 1), cmd)
-	bb.note("Unpacking %s to %s/" % (file, os.getcwd()))
+	bb.note("Unpacking %s to %s/" % (base_path_out(file, data), base_path_out(os.getcwd(), data)))
 	ret = os.system(cmd)
 
 	os.chdir(save_cwd)
classes/packaged-staging.bbclass +1 -1
index 82c1a0d..bd7b9ea 100644
@@ -135,7 +135,7 @@ do_clean_prepend() {
 	pstage_cleanpackage(removepkg, d)
 
 	stagepkg = bb.data.expand("${PSTAGE_PKG}", d)
-	bb.note("Removing staging package %s" % stagepkg)
+	bb.note("Removing staging package %s" % base_path_out(stagepkg, d))
 	os.system('rm -rf ' + stagepkg)
 }
 
classes/patch.bbclass +1 -1
index 075e826..cce8289 100644
@@ -519,7 +519,7 @@ python patch_do_patch() {
 				bb.note("Patch '%s' applies to earlier revisions" % pname)
 				continue
 
-		bb.note("Applying patch '%s' (%s)" % (pname, unpacked))
+		bb.note("Applying patch '%s' (%s)" % (pname, base_path_out(unpacked, d)))
 		try:
 			patchset.Import({"file":unpacked, "remote":url, "strippath": pnum}, True)
 		except:
Signature · 1 line
1.6.0.2
#2 lore raw

Re: [PATCH] Shorten some full paths printed to the user.

Quoted 6 lines · Chris Larson

Adds a base_path_out convenience function, which prepares a full path for display to the user. The initial implementation just makes it relative to ${TOPDIR}. This function is then used for some messages outputted to the user (packaged-staging, patch application, clean, unpack tasks).

Signature · 3 lines
Otavio Salvador                  O.S. Systems
E-mail: [email protected]  http://www.ossystems.com.br
Mobile: +55 53 9981-7854         http://projetos.ossystems.com.br
#3 lore raw

Re: [PATCH] Shorten some full paths printed to the user.

Quoted 10 lines · Otavio Salvador
Quoted 6 lines · Chris Larson

Adds a base_path_out convenience function, which prepares a full path for display to the user. The initial implementation just makes it relative to ${TOPDIR}. This function is then used for some messages outputted to the user (packaged-staging, patch application, clean, unpack tasks).

Yeah, I like the second patch! Not sure about the first one - I personally do have filters on the output messages to highlight and/or notify me about what's going on with the build.

Signature · 1 line
Denys
#4 lore raw
In reply to Chris Larson
Quoted 6 lines · Chris Larson

OpenEmbedded outputs a lot of messages that the user is likely to never care about. We should only output something when it reflects upon their recipe (i.e. unpacking their sources, applying their patches), or is quite significant or unusual.

Signature · 3 lines
Otavio Salvador                  O.S. Systems
E-mail: [email protected]  http://www.ossystems.com.br
Mobile: +55 53 9981-7854         http://projetos.ossystems.com.br
#5 lore raw
Quoted 10 lines · Otavio Salvador
Quoted 6 lines · Chris Larson

OpenEmbedded outputs a lot of messages that the user is likely to never care about. We should only output something when it reflects upon their recipe (i.e. unpacking their sources, applying their patches), or is quite significant or unusual.

Did you mean Acked-by? :)

Signature · 1 line
Denys
#6 lore raw
Quoted 8 lines · Denys Dmytriyenko
Quoted 7 lines · Otavio Salvador
Quoted 5 lines · Chris Larson

OpenEmbedded outputs a lot of messages that the user is likely to never care about. We should only output something when it reflects upon their recipe (i.e. unpacking their sources, applying their patches), or is quite significant or unusual.

Quoted 5 lines
Quoted 3 lines
Quoted 1 line

Did you mean Acked-by? :)

No, since I'm not a core developer :-)

Signature · 3 lines
Otavio Salvador                  O.S. Systems
E-mail: [email protected]  http://www.ossystems.com.br
Mobile: +55 53 9981-7854         http://projetos.ossystems.com.br
#7 lore raw
Quoted 10 lines · Otavio Salvador
Quoted 8 lines · Denys Dmytriyenko
Quoted 7 lines · Otavio Salvador
Quoted 5 lines · Chris Larson

OpenEmbedded outputs a lot of messages that the user is likely to never care about. We should only output something when it reflects upon their recipe (i.e. unpacking their sources, applying their patches), or is quite significant or unusual.

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
Quoted 1 line

Did you mean Acked-by? :)

No, since I'm not a core developer :-)

Signed-off-by is generally used by those involved in the development of the patch, or those through whom the patch has gone in its path upstream, i.e. when it goes through a gatekeeper, afaik. I doubt Acked-by is limited to the core devs.. but I'm talking from a general sign off usage perspective here, not the OE specific usage of it. *shrug*

Signature · 5 lines
Chris Larson
Software Engineer
MontaVista Software, Inc.
Email: clarson AT mvista DOT com
Email: clarson AT kergoth DOT com