[PATCH 00/22] Use bb.fatal() instead of raising FuncFailed

[PATCH 00/22] Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

Motiving quote:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

This seems to be true from looking at the code. exec_func_python() in
bitbake/lib/bb/build.py raises FuncFailed when it catches the
BBHandledException raised by bb.fatal(). Metadata functions that catch
FuncFailed (e.g. in image.bclass) should still work, per the original purpose
of FuncFailed.

The output might change slightly (imo for the better). Here's an example before
the change:

  ERROR: gettext-0.19.8.1-r0 do_fetch: Function failed: Fetcher failure for URL: 'http://ftp.gnu.org/gnu/gettext/gettext-0.19.8.1.tar.gz'. ...
  ERROR: Logfile of failure stored in: /home/ulf/poky/build/tmp/work/i586-poky-linux/gettext/0.19.8.1-r0/temp/log.do_fetch.19715

Here's what it looks like after the change:

  ERROR: gettext-0.19.8.1-r0 do_fetch: Fetcher failure for URL: 'http://ftp.gnu.org/gnu/gettext/gettext-0.19.8.1.tar.gz'. ...
  ERROR: gettext-0.19.8.1-r0 do_fetch: Function failed: base_do_fetch
  ERROR: Logfile of failure stored in: /home/ulf/poky/build/tmp/work/i586-poky-linux/gettext/0.19.8.1-r0/temp/log.do_fetch.19608

I won't get terribly offended if not all of these get merged. Hopefully it
makes people aware that there's a simpler way at least.

There's also a few instances in meta/lib/oe and meta/lib/oeqa that could
probably be replaced, but I got tired and wasn't sure that they're always
called in a context where it'd be safe.

Ulf Magnusson (22):
  base.bbclass: Use bb.fatal() instead of raising FuncFailed
  package_ipk.bbclass: Use bb.fatal() instead of raising FuncFailed
  package_deb.bbclass: Use bb.fatal() instead of raising FuncFailed
  package_rpm.bbclass: Use bb.fatal() instead of raising FuncFailed
  useradd-staticids.bbclass: Use bb.fatal() instead of raising
    FuncFailed
  grub-efi.bbclass: Use bb.fatal() instead of raising FuncFailed
  syslinux.bbclass: Use bb.fatal() instead of raising FuncFailed
  systemd-boot.bbclass: Use bb.fatal() instead of raising FuncFailed
  gummiboot.bbclass: Use bb.fatal() instead of raising FuncFailed
  update-rc.d.bbclass: Use bb.fatal() instead of raising FuncFailed
  license.bbclass: Use bb.fatal() instead of raising FuncFailed
  systemd.bbclass: Use bb.fatal() instead of raising FuncFailed
  gtk-immodules-cache.bbclass: Use bb.fatal() instead of raising
    FuncFailed
  useradd.bbclass: Use bb.fatal() instead of raising FuncFailed
  sstate.bbclass: Use bb.fatal() instead of raising FuncFailed
  chrpath.bbclass: Use bb.fatal() instead of raising FuncFailed
  testsdk.bbclass: Use bb.fatal() instead of raising FuncFailed
  libc-package.bbclass: Use bb.fatal() instead of raising FuncFailed
  package.bbclass: Use bb.fatal() instead of raising FuncFailed
  utility-tasks.bbclass: Use bb.fatal() instead of raising FuncFailed
  testimage.bbclass: Use bb.fatal() instead of raising FuncFailed
  cmake: Use bb.fatal() instead of raising FuncFailed

 meta/classes/base.bbclass                  |  8 ++++----
 meta/classes/chrpath.bbclass               |  3 +--
 meta/classes/grub-efi.bbclass              |  8 ++++----
 meta/classes/gtk-immodules-cache.bbclass   |  2 +-
 meta/classes/gummiboot.bbclass             |  6 +++---
 meta/classes/libc-package.bbclass          |  2 +-
 meta/classes/license.bbclass               |  4 ++--
 meta/classes/package.bbclass               |  2 +-
 meta/classes/package_deb.bbclass           | 10 +++++-----
 meta/classes/package_ipk.bbclass           | 10 +++++-----
 meta/classes/package_rpm.bbclass           |  6 +++---
 meta/classes/sstate.bbclass                |  2 +-
 meta/classes/syslinux.bbclass              |  8 ++++----
 meta/classes/systemd-boot.bbclass          |  6 +++---
 meta/classes/systemd.bbclass               |  3 +--
 meta/classes/testimage.bbclass             |  2 +-
 meta/classes/testsdk.bbclass               |  2 +-
 meta/classes/update-rc.d.bbclass           |  4 ++--
 meta/classes/useradd-staticids.bbclass     |  8 ++++----
 meta/classes/useradd.bbclass               |  2 +-
 meta/classes/utility-tasks.bbclass         |  2 +-
 meta/recipes-devtools/cmake/cmake_3.6.1.bb |  2 +-
 22 files changed, 50 insertions(+), 52 deletions(-)

-- 
2.7.4

[PATCH 01/22] base.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/base.bbclass | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/meta/classes/base.bbclass b/meta/classes/base.bbclass
index 21957d8..024fe43 100644
--- a/meta/classes/base.bbclass
+++ b/meta/classes/base.bbclass
@@ -113,7 +113,7 @@ def get_lic_checksum_file_list(d):
                     continue
                 filelist.append(path + ":" + str(os.path.exists(path)))
         except bb.fetch.MalformedUrl:
-            raise bb.build.FuncFailed(d.getVar('PN', True) + ": LIC_FILES_CHKSUM contains an invalid URL: " + url)
+            bb.fatal(d.getVar('PN', True) + ": LIC_FILES_CHKSUM contains an invalid URL: " + url)
     return " ".join(filelist)
 
 addtask fetch
@@ -131,7 +131,7 @@ python base_do_fetch() {
         fetcher = bb.fetch2.Fetch(src_uri, d)
         fetcher.download()
     except bb.fetch2.BBFetchException as e:
-        raise bb.build.FuncFailed(e)
+        bb.fatal(str(e))
 }
 
 addtask unpack after do_fetch
@@ -152,7 +152,7 @@ python base_do_unpack() {
         fetcher = bb.fetch2.Fetch(src_uri, d)
         fetcher.unpack(d.getVar('WORKDIR', True))
     except bb.fetch2.BBFetchException as e:
-        raise bb.build.FuncFailed(e)
+        bb.fatal(str(e))
 }
 
 def pkgarch_mapping(d):
@@ -671,7 +671,7 @@ python do_cleanall() {
         fetcher = bb.fetch2.Fetch(src_uri, d)
         fetcher.clean()
     except bb.fetch2.BBFetchException as e:
-        raise bb.build.FuncFailed(e)
+        bb.fatal(str(e))
 }
 do_cleanall[nostamp] = "1"
 
-- 
2.7.4

[PATCH 02/22] package_ipk.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/package_ipk.bbclass | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/meta/classes/package_ipk.bbclass b/meta/classes/package_ipk.bbclass
index 930e154..eb00932 100644
--- a/meta/classes/package_ipk.bbclass
+++ b/meta/classes/package_ipk.bbclass
@@ -108,7 +108,7 @@ python do_package_ipk () {
             ctrlfile = open(os.path.join(controldir, 'control'), 'w')
         except OSError:
             bb.utils.unlockfile(lf)
-            raise bb.build.FuncFailed("unable to open control file for writing.")
+            bb.fatal("unable to open control file for writing")
 
         fields = []
         pe = d.getVar('PKGE', True)
@@ -162,7 +162,7 @@ python do_package_ipk () {
             (type, value, traceback) = sys.exc_info()
             ctrlfile.close()
             bb.utils.unlockfile(lf)
-            raise bb.build.FuncFailed("Missing field for ipk generation: %s" % value)
+            bb.fatal("Missing field for ipk generation: %s" % value)
         # more fields
 
         custom_fields_chunk = get_package_additional_metadata("ipk", localdata)
@@ -226,7 +226,7 @@ python do_package_ipk () {
                 scriptfile = open(os.path.join(controldir, script), 'w')
             except OSError:
                 bb.utils.unlockfile(lf)
-                raise bb.build.FuncFailed("unable to open %s script file for writing." % script)
+                bb.fatal("unable to open %s script file for writing" % script)
             scriptfile.write(scriptvar)
             scriptfile.close()
             os.chmod(os.path.join(controldir, script), 0o755)
@@ -237,7 +237,7 @@ python do_package_ipk () {
                 conffiles = open(os.path.join(controldir, 'conffiles'), 'w')
             except OSError:
                 bb.utils.unlockfile(lf)
-                raise bb.build.FuncFailed("unable to open conffiles for writing.")
+                bb.fatal("unable to open conffiles for writing")
             for f in conffiles_str.split():
                 if os.path.exists(oe.path.join(root, f)):
                     conffiles.write('%s\n' % f)
@@ -248,7 +248,7 @@ python do_package_ipk () {
                                                           d.getVar("OPKGBUILDCMD", True), pkg, pkgoutdir), shell=True)
         if ret != 0:
             bb.utils.unlockfile(lf)
-            raise bb.build.FuncFailed("opkg-build execution failed")
+            bb.fatal("opkg-build execution failed")
 
         if d.getVar('IPK_SIGN_PACKAGES', True) == '1':
             ipkver = "%s-%s" % (d.getVar('PKGV', True), d.getVar('PKGR', True))
-- 
2.7.4

[PATCH 03/22] package_deb.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/package_deb.bbclass | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/meta/classes/package_deb.bbclass b/meta/classes/package_deb.bbclass
index 4e5dc12..fb6034c 100644
--- a/meta/classes/package_deb.bbclass
+++ b/meta/classes/package_deb.bbclass
@@ -126,7 +126,7 @@ python do_package_deb () {
             ctrlfile = codecs.open(os.path.join(controldir, 'control'), 'w', 'utf-8')
         except OSError:
             bb.utils.unlockfile(lf)
-            raise bb.build.FuncFailed("unable to open control file for writing.")
+            bb.fatal("unable to open control file for writing")
 
         fields = []
         pe = d.getVar('PKGE', True)
@@ -197,7 +197,7 @@ python do_package_deb () {
             (type, value, traceback) = sys.exc_info()
             bb.utils.unlockfile(lf)
             ctrlfile.close()
-            raise bb.build.FuncFailed("Missing field for deb generation: %s" % value)
+            bb.fatal("Missing field for deb generation: %s" % value)
 
         # more fields
 
@@ -277,7 +277,7 @@ python do_package_deb () {
                 scriptfile = open(os.path.join(controldir, script), 'w')
             except OSError:
                 bb.utils.unlockfile(lf)
-                raise bb.build.FuncFailed("unable to open %s script file for writing." % script)
+                bb.fatal("unable to open %s script file for writing" % script)
 
             if scriptvar.startswith("#!"):
                 pos = scriptvar.find("\n") + 1
@@ -301,7 +301,7 @@ python do_package_deb () {
                 conffiles = open(os.path.join(controldir, 'conffiles'), 'w')
             except OSError:
                 bb.utils.unlockfile(lf)
-                raise bb.build.FuncFailed("unable to open conffiles for writing.")
+                bb.fatal("unable to open conffiles for writing")
             for f in conffiles_str.split():
                 if os.path.exists(oe.path.join(root, f)):
                     conffiles.write('%s\n' % f)
@@ -311,7 +311,7 @@ python do_package_deb () {
         ret = subprocess.call("PATH=\"%s\" dpkg-deb -b %s %s" % (localdata.getVar("PATH", True), root, pkgoutdir), shell=True)
         if ret != 0:
             bb.utils.unlockfile(lf)
-            raise bb.build.FuncFailed("dpkg-deb execution failed")
+            bb.fatal("dpkg-deb execution failed")
 
         cleanupcontrol(root)
         bb.utils.unlockfile(lf)
-- 
2.7.4

[PATCH 04/22] package_rpm.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/package_rpm.bbclass | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/meta/classes/package_rpm.bbclass b/meta/classes/package_rpm.bbclass
index f9398a9..c431545 100644
--- a/meta/classes/package_rpm.bbclass
+++ b/meta/classes/package_rpm.bbclass
@@ -58,7 +58,7 @@ def write_rpm_perfiledata(srcname, d):
     try:
         dependsfile = open(outdepends, 'w')
     except OSError:
-        raise bb.build.FuncFailed("unable to open spec file for writing.")
+        bb.fatal("unable to open spec file for writing")
 
     dump_filerdeps('RDEPENDS', dependsfile, d)
 
@@ -71,7 +71,7 @@ def write_rpm_perfiledata(srcname, d):
     try:
         providesfile = open(outprovides, 'w')
     except OSError:
-        raise bb.build.FuncFailed("unable to open spec file for writing.")
+        bb.fatal("unable to open spec file for writing")
 
     dump_filerdeps('RPROVIDES', providesfile, d)
 
@@ -617,7 +617,7 @@ python write_specfile () {
     try:
         specfile = open(outspecfile, 'w')
     except OSError:
-        raise bb.build.FuncFailed("unable to open spec file for writing.")
+        bb.fatal("unable to open spec file for writing")
 
     # RPMSPEC_PREAMBLE is a way to add arbitrary text to the top
     # of the generated spec file
-- 
2.7.4

[PATCH 05/22] useradd-staticids.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/useradd-staticids.bbclass | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/meta/classes/useradd-staticids.bbclass b/meta/classes/useradd-staticids.bbclass
index 46d4a4b..afb580a 100644
--- a/meta/classes/useradd-staticids.bbclass
+++ b/meta/classes/useradd-staticids.bbclass
@@ -16,7 +16,7 @@ def update_useradd_static_config(d):
             error(message)
 
         def error(self, message):
-            raise bb.build.FuncFailed(message)
+            bb.fatal(message)
 
     def list_extend(iterable, length, obj = None):
         """Ensure that iterable is the specified length by extending with obj
@@ -54,7 +54,7 @@ def update_useradd_static_config(d):
         # For backwards compatibility we accept "1" in addition to "error"
         if d.getVar('USERADD_ERROR_DYNAMIC', True) == 'error' or d.getVar('USERADD_ERROR_DYNAMIC', True) == '1':
             #bb.error("Skipping recipe %s, package %s which adds %sname %s does not have a static ID defined." % (d.getVar('PN', True),  pkg, type, id))
-            raise bb.build.FuncFailed("%s - %s: %sname %s does not have a static ID defined." % (d.getVar('PN', True), pkg, type, id))
+            bb.fatal("%s - %s: %sname %s does not have a static ID defined." % (d.getVar('PN', True), pkg, type, id))
         elif d.getVar('USERADD_ERROR_DYNAMIC', True) == 'warn':
             bb.warn("%s - %s: %sname %s does not have a static ID defined." % (d.getVar('PN', True), pkg, type, id))
 
@@ -107,7 +107,7 @@ def update_useradd_static_config(d):
             try:
                 uaargs = parser.parse_args(re.split('''[ \t]+(?=(?:[^'"]|'[^']*'|"[^"]*")*$)''', param))
             except:
-                raise bb.build.FuncFailed("%s: Unable to parse arguments for USERADD_PARAM_%s: '%s'" % (d.getVar('PN', True), pkg, param))
+                bb.fatal("%s: Unable to parse arguments for USERADD_PARAM_%s: '%s'" % (d.getVar('PN', True), pkg, param))
 
             # Read all passwd files specified in USERADD_UID_TABLES or files/passwd
             # Use the standard passwd layout:
@@ -242,7 +242,7 @@ def update_useradd_static_config(d):
                 # If we're processing multiple lines, we could have left over values here...
                 gaargs = parser.parse_args(re.split('''[ \t]+(?=(?:[^'"]|'[^']*'|"[^"]*")*$)''', param))
             except:
-                raise bb.build.FuncFailed("%s: Unable to parse arguments for GROUPADD_PARAM_%s: '%s'" % (d.getVar('PN', True), pkg, param))
+                bb.fatal("%s: Unable to parse arguments for GROUPADD_PARAM_%s: '%s'" % (d.getVar('PN', True), pkg, param))
 
             # Read all group files specified in USERADD_GID_TABLES or files/group
             # Use the standard group layout:
-- 
2.7.4

[PATCH 06/22] grub-efi.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/grub-efi.bbclass | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/meta/classes/grub-efi.bbclass b/meta/classes/grub-efi.bbclass
index 26ba8ce..17417ba 100644
--- a/meta/classes/grub-efi.bbclass
+++ b/meta/classes/grub-efi.bbclass
@@ -90,12 +90,12 @@ python build_efi_cfg() {
 
     cfile = d.getVar('GRUB_CFG', True)
     if not cfile:
-        raise bb.build.FuncFailed('Unable to read GRUB_CFG')
+        bb.fatal('Unable to read GRUB_CFG')
 
     try:
          cfgfile = open(cfile, 'w')
     except OSError:
-        raise bb.build.FuncFailed('Unable to open %s' % (cfile))
+        bb.fatal('Unable to open %s' % cfile)
 
     cfgfile.write('# Automatically created by OE\n')
 
@@ -114,7 +114,7 @@ python build_efi_cfg() {
 
     root = d.getVar('GRUB_ROOT', True)
     if not root:
-        raise bb.build.FuncFailed('GRUB_ROOT not defined')
+        bb.fatal('GRUB_ROOT not defined')
 
     if gfxserial == "1":
         btypes = [ [ " graphics console", "" ],
@@ -127,7 +127,7 @@ python build_efi_cfg() {
 
         overrides = localdata.getVar('OVERRIDES', True)
         if not overrides:
-            raise bb.build.FuncFailed('OVERRIDES not defined')
+            bb.fatal('OVERRIDES not defined')
 
         for btype in btypes:
             localdata.setVar('OVERRIDES', label + ':' + overrides)
-- 
2.7.4

[PATCH 07/22] syslinux.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/syslinux.bbclass | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/meta/classes/syslinux.bbclass b/meta/classes/syslinux.bbclass
index 9b0c2c7..7778fd7 100644
--- a/meta/classes/syslinux.bbclass
+++ b/meta/classes/syslinux.bbclass
@@ -100,12 +100,12 @@ python build_syslinux_cfg () {
 
     cfile = d.getVar('SYSLINUX_CFG', True)
     if not cfile:
-        raise bb.build.FuncFailed('Unable to read SYSLINUX_CFG')
+        bb.fatal('Unable to read SYSLINUX_CFG')
 
     try:
         cfgfile = open(cfile, 'w')
     except OSError:
-        raise bb.build.FuncFailed('Unable to open %s' % (cfile))
+        bb.fatal('Unable to open %s' % cfile)
 
     cfgfile.write('# Automatically created by OE\n')
 
@@ -160,7 +160,7 @@ python build_syslinux_cfg () {
 
         overrides = localdata.getVar('OVERRIDES', True)
         if not overrides:
-            raise bb.build.FuncFailed('OVERRIDES not defined')
+            bb.fatal('OVERRIDES not defined')
 
         localdata.setVar('OVERRIDES', label + ':' + overrides)
         bb.data.update_data(localdata)
@@ -172,7 +172,7 @@ python build_syslinux_cfg () {
 
         root= d.getVar('SYSLINUX_ROOT', True)
         if not root:
-            raise bb.build.FuncFailed('SYSLINUX_ROOT not defined')
+            bb.fatal('SYSLINUX_ROOT not defined')
 
         for btype in btypes:
             cfgfile.write('LABEL %s%s\nKERNEL /vmlinuz\n' % (btype[0], label))
-- 
2.7.4

[PATCH 08/22] systemd-boot.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/systemd-boot.bbclass | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/meta/classes/systemd-boot.bbclass b/meta/classes/systemd-boot.bbclass
index b550b61..05244c7 100644
--- a/meta/classes/systemd-boot.bbclass
+++ b/meta/classes/systemd-boot.bbclass
@@ -77,7 +77,7 @@ python build_efi_cfg() {
     try:
          cfgfile = open(cfile, 'w')
     except OSError:
-        raise bb.build.FuncFailed('Unable to open %s' % (cfile))
+        bb.fatal('Unable to open %s' % cfile)
 
     cfgfile.write('# Automatically created by OE\n')
     cfgfile.write('default %s\n' % (labels.split()[0]))
@@ -93,14 +93,14 @@ python build_efi_cfg() {
 
         overrides = localdata.getVar('OVERRIDES', True)
         if not overrides:
-            raise bb.build.FuncFailed('OVERRIDES not defined')
+            bb.fatal('OVERRIDES not defined')
 
         entryfile = "%s/%s.conf" % (s, label)
         d.appendVar("SYSTEMD_BOOT_ENTRIES", " " + entryfile)
         try:
             entrycfg = open(entryfile, "w")
         except OSError:
-            raise bb.build.FuncFailed('Unable to open %s' % (entryfile))
+            bb.fatal('Unable to open %s' % entryfile)
         localdata.setVar('OVERRIDES', label + ':' + overrides)
         bb.data.update_data(localdata)
 
-- 
2.7.4

[PATCH 09/22] gummiboot.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/gummiboot.bbclass | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/meta/classes/gummiboot.bbclass b/meta/classes/gummiboot.bbclass
index e6eba17..4f2dea6 100644
--- a/meta/classes/gummiboot.bbclass
+++ b/meta/classes/gummiboot.bbclass
@@ -74,7 +74,7 @@ python build_efi_cfg() {
     try:
          cfgfile = open(cfile, 'w')
     except OSError:
-        raise bb.build.FuncFailed('Unable to open %s' % (cfile))
+        bb.fatal('Unable to open %s' % cfile)
 
     cfgfile.write('# Automatically created by OE\n')
     cfgfile.write('default %s\n' % (labels.split()[0]))
@@ -90,14 +90,14 @@ python build_efi_cfg() {
 
         overrides = localdata.getVar('OVERRIDES', True)
         if not overrides:
-            raise bb.build.FuncFailed('OVERRIDES not defined')
+            bb.fatal('OVERRIDES not defined')
 
         entryfile = "%s/%s.conf" % (s, label)
         d.appendVar("GUMMIBOOT_ENTRIES", " " + entryfile)
         try:
             entrycfg = open(entryfile, "w")
         except OSError:
-            raise bb.build.FuncFailed('Unable to open %s' % (entryfile))
+            bb.fatal('Unable to open %s' % entryfile)
         localdata.setVar('OVERRIDES', label + ':' + overrides)
         bb.data.update_data(localdata)
 
-- 
2.7.4

[PATCH 10/22] update-rc.d.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/update-rc.d.bbclass | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/meta/classes/update-rc.d.bbclass b/meta/classes/update-rc.d.bbclass
index 82b8024..dfef2a2 100644
--- a/meta/classes/update-rc.d.bbclass
+++ b/meta/classes/update-rc.d.bbclass
@@ -57,9 +57,9 @@ fi
 def update_rc_after_parse(d):
     if d.getVar('INITSCRIPT_PACKAGES', False) == None:
         if d.getVar('INITSCRIPT_NAME', False) == None:
-            raise bb.build.FuncFailed("%s inherits update-rc.d but doesn't set INITSCRIPT_NAME" % d.getVar('FILE', False))
+            bb.fatal("%s inherits update-rc.d but doesn't set INITSCRIPT_NAME" % d.getVar('FILE', False))
         if d.getVar('INITSCRIPT_PARAMS', False) == None:
-            raise bb.build.FuncFailed("%s inherits update-rc.d but doesn't set INITSCRIPT_PARAMS" % d.getVar('FILE', False))
+            bb.fatal("%s inherits update-rc.d but doesn't set INITSCRIPT_PARAMS" % d.getVar('FILE', False))
 
 python __anonymous() {
     update_rc_after_parse(d)
-- 
2.7.4

[PATCH 11/22] license.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/license.bbclass | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/meta/classes/license.bbclass b/meta/classes/license.bbclass
index ad12db4..da4fc3e 100644
--- a/meta/classes/license.bbclass
+++ b/meta/classes/license.bbclass
@@ -465,7 +465,7 @@ def find_license_files(d):
             pass
 
     if not generic_directory:
-        raise bb.build.FuncFailed("COMMON_LICENSE_DIR is unset. Please set this in your distro config")
+        bb.fatal("COMMON_LICENSE_DIR is unset. Please set this in your distro config")
 
     if not lic_files:
         # No recipe should have an invalid license file. This is checked else
@@ -477,7 +477,7 @@ def find_license_files(d):
         try:
             (type, host, path, user, pswd, parm) = bb.fetch.decodeurl(url)
         except bb.fetch.MalformedUrl:
-            raise bb.build.FuncFailed("%s: LIC_FILES_CHKSUM contains an invalid URL:  %s" % (d.getVar('PF', True), url))
+            bb.fatal("%s: LIC_FILES_CHKSUM contains an invalid URL:  %s" % (d.getVar('PF', True), url))
         # We want the license filename and path
         chksum = parm['md5'] if 'md5' in parm else parm['sha256']
         lic_chksums[path] = chksum
-- 
2.7.4

[PATCH 12/22] systemd.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/systemd.bbclass | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/meta/classes/systemd.bbclass b/meta/classes/systemd.bbclass
index db7873f..d56c760 100644
--- a/meta/classes/systemd.bbclass
+++ b/meta/classes/systemd.bbclass
@@ -165,8 +165,7 @@ python systemd_populate_packages() {
                 if path_found != '':
                     systemd_add_files_and_parse(pkg_systemd, path_found, service, keys)
                 else:
-                    raise bb.build.FuncFailed("SYSTEMD_SERVICE_%s value %s does not exist" % \
-                        (pkg_systemd, service))
+                    bb.fatal("SYSTEMD_SERVICE_%s value %s does not exist" % (pkg_systemd, service))
 
     # Run all modifications once when creating package
     if os.path.exists(d.getVar("D", True)):
-- 
2.7.4

[PATCH 13/22] gtk-immodules-cache.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/gtk-immodules-cache.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/classes/gtk-immodules-cache.bbclass b/meta/classes/gtk-immodules-cache.bbclass
index 3a4634c..ebbc9de 100644
--- a/meta/classes/gtk-immodules-cache.bbclass
+++ b/meta/classes/gtk-immodules-cache.bbclass
@@ -84,6 +84,6 @@ python __anonymous() {
         gtkimmodules_check = d.getVar('GTKIMMODULES_PACKAGES', False)
         if not gtkimmodules_check:
             bb_filename = d.getVar('FILE', False)
-            raise bb.build.FuncFailed("ERROR: %s inherits gtk-immodules-cache but doesn't set GTKIMMODULES_PACKAGES" % bb_filename)
+            bb.fatal("ERROR: %s inherits gtk-immodules-cache but doesn't set GTKIMMODULES_PACKAGES" % bb_filename)
 }
 
-- 
2.7.4

[PATCH 14/22] useradd.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/useradd.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/classes/useradd.bbclass b/meta/classes/useradd.bbclass
index bf62ada..3cff08e 100644
--- a/meta/classes/useradd.bbclass
+++ b/meta/classes/useradd.bbclass
@@ -171,7 +171,7 @@ def update_useradd_after_parse(d):
     useradd_packages = d.getVar('USERADD_PACKAGES', True)
 
     if not useradd_packages:
-        raise bb.build.FuncFailed("%s inherits useradd but doesn't set USERADD_PACKAGES" % d.getVar('FILE', False))
+        bb.fatal("%s inherits useradd but doesn't set USERADD_PACKAGES" % d.getVar('FILE', False))
 
     for pkg in useradd_packages.split():
         if not d.getVar('USERADD_PARAM_%s' % pkg, True) and not d.getVar('GROUPADD_PARAM_%s' % pkg, True) and not d.getVar('GROUPMEMS_PARAM_%s' % pkg, True):
-- 
2.7.4

[PATCH 15/22] sstate.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/sstate.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/classes/sstate.bbclass b/meta/classes/sstate.bbclass
index ac9d77d..6f0c791 100644
--- a/meta/classes/sstate.bbclass
+++ b/meta/classes/sstate.bbclass
@@ -658,7 +658,7 @@ def sstate_setscene(d):
     shared_state = sstate_state_fromvars(d)
     accelerate = sstate_installpkg(shared_state, d)
     if not accelerate:
-        raise bb.build.FuncFailed("No suitable staging package found")
+        bb.fatal("No suitable staging package found")
 
 python sstate_task_prefunc () {
     shared_state = sstate_state_fromvars(d)
-- 
2.7.4

[PATCH 16/22] chrpath.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/chrpath.bbclass | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/meta/classes/chrpath.bbclass b/meta/classes/chrpath.bbclass
index 72e8a12..3b5cd37 100644
--- a/meta/classes/chrpath.bbclass
+++ b/meta/classes/chrpath.bbclass
@@ -39,8 +39,7 @@ def process_file_linux(cmd, fpath, rootdir, baseprefix, tmpdir, d):
         p = sub.Popen([cmd, '-r', args, fpath],stdout=sub.PIPE,stderr=sub.PIPE)
         out, err = p.communicate()
         if p.returncode != 0:
-            bb.error("%s: chrpath command failed with exit code %d:\n%s%s" % (d.getVar('PN', True), p.returncode, out, err))
-            raise bb.build.FuncFailed
+            bb.fatal("%s: chrpath command failed with exit code %d:\n%s%s" % (d.getVar('PN', True), p.returncode, out, err))
 
 def process_file_darwin(cmd, fpath, rootdir, baseprefix, tmpdir, d):
     import subprocess as sub
-- 
2.7.4

[PATCH 17/22] testsdk.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/testsdk.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/classes/testsdk.bbclass b/meta/classes/testsdk.bbclass
index 0b8716e..77c9203 100644
--- a/meta/classes/testsdk.bbclass
+++ b/meta/classes/testsdk.bbclass
@@ -44,7 +44,7 @@ def run_test_context(CTestContext, d, testdir, tcname, pn, *args):
                 msg += " (skipped=%d)" % skipped
             bb.plain(msg)
         else:
-            raise bb.build.FuncFailed("%s - FAILED - check the task log and the commands log" % pn )
+            bb.fatal("%s - FAILED - check the task log and the commands log" % pn)
 
 def testsdk_main(d):
     import os
-- 
2.7.4

[PATCH 18/22] libc-package.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/libc-package.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/classes/libc-package.bbclass b/meta/classes/libc-package.bbclass
index 8c99d61..4d17ff9 100644
--- a/meta/classes/libc-package.bbclass
+++ b/meta/classes/libc-package.bbclass
@@ -260,7 +260,7 @@ python package_do_split_gconvs () {
                 localedef_opts = locale_arch_options[target_arch]
             else:
                 bb.error("locale_arch_options not found for target_arch=" + target_arch)
-                raise bb.build.FuncFailed("unknown arch:" + target_arch + " for locale_arch_options")
+                bb.fatal("unknown arch:" + target_arch + " for locale_arch_options")
 
             localedef_opts += " --force  --no-archive --prefix=%s \
                 --inputfile=%s/%s/i18n/locales/%s --charmap=%s %s/%s" \
-- 
2.7.4

[PATCH 19/22] package.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/package.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/classes/package.bbclass b/meta/classes/package.bbclass
index a9ca141..a6f0a7a 100644
--- a/meta/classes/package.bbclass
+++ b/meta/classes/package.bbclass
@@ -1171,7 +1171,7 @@ python populate_packages () {
                 continue
             ret = bb.utils.copyfile(file, fpath)
             if ret is False or ret == 0:
-                raise bb.build.FuncFailed("File population failed")
+                bb.fatal("File population failed")
 
         # Check if symlink paths exist
         for file in symlink_paths:
-- 
2.7.4

[PATCH 20/22] utility-tasks.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/utility-tasks.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/classes/utility-tasks.bbclass b/meta/classes/utility-tasks.bbclass
index 7bc584a..7ba56e2 100644
--- a/meta/classes/utility-tasks.bbclass
+++ b/meta/classes/utility-tasks.bbclass
@@ -47,7 +47,7 @@ python do_checkuri() {
         fetcher = bb.fetch2.Fetch(src_uri, d)
         fetcher.checkstatus()
     except bb.fetch2.BBFetchException as e:
-        raise bb.build.FuncFailed(e)
+        bb.fatal(str(e))
 }
 
 addtask checkuriall after do_checkuri
-- 
2.7.4

[PATCH 21/22] testimage.bbclass: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/classes/testimage.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/classes/testimage.bbclass b/meta/classes/testimage.bbclass
index a908f92..5ddbecb 100644
--- a/meta/classes/testimage.bbclass
+++ b/meta/classes/testimage.bbclass
@@ -164,7 +164,7 @@ def testimage_main(d):
                 msg += " (skipped=%d)" % skipped
             bb.plain(msg)
         else:
-            raise bb.build.FuncFailed("%s - FAILED - check the task log and the ssh log" % pn )
+            bb.fatal("%s - FAILED - check the task log and the ssh log" % pn)
     finally:
         signal.signal(signal.SIGTERM, tc.origsigtermhandler)
         target.stop()
-- 
2.7.4

[PATCH 22/22] cmake: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

This sets a good example and avoids unnecessarily contributing to
perceived complexity and cargo culting.

Motivating quote below:

< kergoth> the *original* intent was for the function/task to error via
           whatever appropriate means, bb.fatal, whatever, and
           funcfailed was what you'd catch if you were calling
           exec_func/exec_task. that is, it's what those functions
           raise, not what metadata functions should be raising
< kergoth> it didn't end up being used that way
< kergoth> but there's really never a reason to raise it yourself

FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
argument, which also shows that the original purpose got lost.

Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
 meta/recipes-devtools/cmake/cmake_3.6.1.bb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/recipes-devtools/cmake/cmake_3.6.1.bb b/meta/recipes-devtools/cmake/cmake_3.6.1.bb
index 67ba9c4..850d6de 100644
--- a/meta/recipes-devtools/cmake/cmake_3.6.1.bb
+++ b/meta/recipes-devtools/cmake/cmake_3.6.1.bb
@@ -14,7 +14,7 @@ python () {
     docdir=d.getVar("docdir", True)
 
     if not docdir.startswith(prefix):
-        raise bb.build.FuncFailed('docdir must contain prefix as its prefix')
+        bb.fatal('docdir must contain prefix as its prefix')
 
     docdir_stripped = docdir[len(prefix):]
     if len(docdir_stripped) > 0 and docdir_stripped[0] == '/':
-- 
2.7.4

Re: [PATCH 22/22] cmake: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

On Sat, Oct 1, 2016 at 4:47 AM, Ulf Magnusson <ulfalizer@gmail.com> wrote:
> This sets a good example and avoids unnecessarily contributing to
> perceived complexity and cargo culting.
>
> Motivating quote below:
>
> < kergoth> the *original* intent was for the function/task to error via
>            whatever appropriate means, bb.fatal, whatever, and
>            funcfailed was what you'd catch if you were calling
>            exec_func/exec_task. that is, it's what those functions
>            raise, not what metadata functions should be raising
> < kergoth> it didn't end up being used that way
> < kergoth> but there's really never a reason to raise it yourself
>
> FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
> argument, which also shows that the original purpose got lost.
>
> Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
> ---
>  meta/recipes-devtools/cmake/cmake_3.6.1.bb | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/meta/recipes-devtools/cmake/cmake_3.6.1.bb b/meta/recipes-devtools/cmake/cmake_3.6.1.bb
> index 67ba9c4..850d6de 100644
> --- a/meta/recipes-devtools/cmake/cmake_3.6.1.bb
> +++ b/meta/recipes-devtools/cmake/cmake_3.6.1.bb
> @@ -14,7 +14,7 @@ python () {
>      docdir=d.getVar("docdir", True)
>
>      if not docdir.startswith(prefix):
> -        raise bb.build.FuncFailed('docdir must contain prefix as its prefix')
> +        bb.fatal('docdir must contain prefix as its prefix')
>
>      docdir_stripped = docdir[len(prefix):]
>      if len(docdir_stripped) > 0 and docdir_stripped[0] == '/':
> --
> 2.7.4
>

Thanks for merging these!

There's now some documentation that mentions FuncFailed as well:
https://www.yoctoproject.org/docs/2.2/bitbake-user-manual/bitbake-user-manual.html#bitbake-style-python-functions-versus-python-functions

Cheers,
Ulf

Re: [PATCH 22/22] cmake: Use bb.fatal() instead of raising FuncFailed

[Richard Purdie]

On Wed, 2016-10-05 at 22:26 +0200, Ulf Magnusson wrote:
> On Sat, Oct 1, 2016 at 4:47 AM, Ulf Magnusson <ulfalizer@gmail.com>
> wrote:
> > 
> > This sets a good example and avoids unnecessarily contributing to
> > perceived complexity and cargo culting.
> > 
> > Motivating quote below:
> > 
> > < kergoth> the *original* intent was for the function/task to error
> > via
> >            whatever appropriate means, bb.fatal, whatever, and
> >            funcfailed was what you'd catch if you were calling
> >            exec_func/exec_task. that is, it's what those functions
> >            raise, not what metadata functions should be raising
> > < kergoth> it didn't end up being used that way
> > < kergoth> but there's really never a reason to raise it yourself
> > 
> > FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
> > argument, which also shows that the original purpose got lost.
> > 
> > Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
> > ---
> >  meta/recipes-devtools/cmake/cmake_3.6.1.bb | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/meta/recipes-devtools/cmake/cmake_3.6.1.bb
> > b/meta/recipes-devtools/cmake/cmake_3.6.1.bb
> > index 67ba9c4..850d6de 100644
> > --- a/meta/recipes-devtools/cmake/cmake_3.6.1.bb
> > +++ b/meta/recipes-devtools/cmake/cmake_3.6.1.bb
> > @@ -14,7 +14,7 @@ python () {
> >      docdir=d.getVar("docdir", True)
> > 
> >      if not docdir.startswith(prefix):
> > -        raise bb.build.FuncFailed('docdir must contain prefix as
> > its prefix')
> > +        bb.fatal('docdir must contain prefix as its prefix')
> > 
> >      docdir_stripped = docdir[len(prefix):]
> >      if len(docdir_stripped) > 0 and docdir_stripped[0] == '/':
> > --
> > 2.7.4
> > 
> Thanks for merging these!
> 
> There's now some documentation that mentions FuncFailed as well:
> https://www.yoctoproject.org/docs/2.2/bitbake-user-manual/bitbake-use
> r-manual.html#bitbake-style-python-functions-versus-python-functions

It did break one of the tests in oe-selftest but we've fixed that...

Cheers,

Richard

Re: [PATCH 22/22] cmake: Use bb.fatal() instead of raising FuncFailed

[Ulf Magnusson]

On Thu, Oct 6, 2016 at 12:15 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
> On Wed, 2016-10-05 at 22:26 +0200, Ulf Magnusson wrote:
>> On Sat, Oct 1, 2016 at 4:47 AM, Ulf Magnusson <ulfalizer@gmail.com>
>> wrote:
>> >
>> > This sets a good example and avoids unnecessarily contributing to
>> > perceived complexity and cargo culting.
>> >
>> > Motivating quote below:
>> >
>> > < kergoth> the *original* intent was for the function/task to error
>> > via
>> >            whatever appropriate means, bb.fatal, whatever, and
>> >            funcfailed was what you'd catch if you were calling
>> >            exec_func/exec_task. that is, it's what those functions
>> >            raise, not what metadata functions should be raising
>> > < kergoth> it didn't end up being used that way
>> > < kergoth> but there's really never a reason to raise it yourself
>> >
>> > FuncFailed.__init__ takes a 'name' argument rather than a 'msg'
>> > argument, which also shows that the original purpose got lost.
>> >
>> > Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
>> > ---
>> >  meta/recipes-devtools/cmake/cmake_3.6.1.bb | 2 +-
>> >  1 file changed, 1 insertion(+), 1 deletion(-)
>> >
>> > diff --git a/meta/recipes-devtools/cmake/cmake_3.6.1.bb
>> > b/meta/recipes-devtools/cmake/cmake_3.6.1.bb
>> > index 67ba9c4..850d6de 100644
>> > --- a/meta/recipes-devtools/cmake/cmake_3.6.1.bb
>> > +++ b/meta/recipes-devtools/cmake/cmake_3.6.1.bb
>> > @@ -14,7 +14,7 @@ python () {
>> >      docdir=d.getVar("docdir", True)
>> >
>> >      if not docdir.startswith(prefix):
>> > -        raise bb.build.FuncFailed('docdir must contain prefix as
>> > its prefix')
>> > +        bb.fatal('docdir must contain prefix as its prefix')
>> >
>> >      docdir_stripped = docdir[len(prefix):]
>> >      if len(docdir_stripped) > 0 and docdir_stripped[0] == '/':
>> > --
>> > 2.7.4
>> >
>> Thanks for merging these!
>>
>> There's now some documentation that mentions FuncFailed as well:
>> https://www.yoctoproject.org/docs/2.2/bitbake-user-manual/bitbake-use
>> r-manual.html#bitbake-style-python-functions-versus-python-functions
>
> It did break one of the tests in oe-selftest but we've fixed that...
>
> Cheers,
>
> Richard

Sorry about that. I wasn't even aware of oe-selftest until now.

Yeah... the format for fetch failures changed slightly.

Cheers,
Ulf