However some existing functions had names which shadowed existing
functions in the List module, so I had to rename them:
assoc -> List.assoc_lbl
append -> List.push_back_list
prepend -> List.push_front_list
This is an extension of the previous commit.
When caching all the templates, use the architecture of each template,
instead of the architecture passed as --arch (or the host architecture,
as default). This way, the right destination filename will be used.
Fixes commit b1cf6246f3.
Thanks to: Erik Skultety.
virt-builder-repository allows users to easily create or update
a virt-builder source repository out of disk images. The tool can
be run in either interactive or automated mode.
get_index now gets a new template parameter. Setting it to true will
make the index parsing less picky about missing important data. This
can be used to parse a partial index file.
Change Index.arch to the type (Arch of string | GuessedArch of string).
In a future commit, the index parser will allow arch not to be set
for some cases. Thus arch value will be guessed by inspecting the
image. However we need to distinguish between a set value and a guessed
one. Using this new type will help it:
match arch with
| Arch s -> (* This is a set value *)
| GuessedArch s -> (* This is a guessed value *)
Every .ml file now needs a corresponding .mli file so that
dependencies are generated correctly. The check-mli.sh script picks
up the problem:
FAIL: check-mli.sh
==================
./builder/osinfo_config.ml: missing ./builder/osinfo_config.mli
FAIL check-mli.sh (exit status: 1)
Fixes commit 46abca0ec3
and commit 02184f55f9.
The immediate issue is with Fedora/ppc64 and /ppc64le which currently
use extended partitions, breaking the virt-builder ‘--size’ parameter,
eg:
$ virt-builder --arch ppc64le fedora-26 --size 20G
...
[ 21.6] Resizing (using virt-resize) to expand the disk to 20.0G
virt-resize: error: /dev/sda5: partition not found in the source disk image
(this error came from '--expand' option on the command line). Try running
this command: virt-filesystems --partitions --long -a /var/tmp/vbf67b8c.img
However more generally MBR is broken and should die. GPT is supported
by all modern virtual bootloaders, so just default to it.
Notes:
* This is different from mandating a UEFI bootloader.
* I am not planning to rebuild any existing images except the
F26 ppc64 & ppc64le ones.
Add a function to properly write virt-builder source index entries.
Note that this function is very similar to Index.print_entry that is
meant for debugging purposes.
Add a simple OCaml-based implementation of reader of the osinfo-db:
the only interface is an iterator that invokes an user-supplied
function with each XML file found.
This implementation behaves like the current C implementation, and
still supports the old libosinfo db.
[RWMJ: Fixed trailing whitespace]
Inspired by ocaml-extlib, introduce a module for handling option
types.
We already had the ‘may’ function (which becomes ‘Option.may’). This
adds also ‘Option.map’ (unused), and ‘Option.default’ functions.
Note this does *not* introduce the unsafe ‘Option.get’ function from
extlib.
We reimplemented some functions which can now be found in the OCaml
stdlib since 4.01 (or earlier). The functions I have dropped are:
- String.map
- |>
- iteri (replaced by List.iteri)
- mapi (replaced by List.mapi)
Note that our definition of iteri was slightly wrong: the type of the
function parameter was too wide, allowing (int -> 'a -> 'b) instead of
(int -> 'a -> unit).
I also added this new function to the Std_utils.String module as an
export from stdlib String:
- String.iteri
Thanks: Pino Toscano
Change the Planner.plan function so it returns an optional type. This
means it no longer raises Failure "plan" on error, so we can both
force the caller to deal with the error case and avoid Warning 52.
This avoids warning 52 in OCaml code such as:
try URI.parse_uri arg
with Invalid_argument "URI.parse_uri" -> ...
which prints:
Warning 52: Code should not depend on the actual values of
this constructor's arguments. They are only for information
and may change in future versions. (See manual section 8.5)
In the long term we need to change fish/uri.c so that we can throw
proper errors.
If you have a struct containing ‘field’, eg:
type t = { field : int }
then previously to pattern-match on this type, eg. in function
parameters, you had to write:
let f { field = field } =
(* ... use field ... *)
In OCaml >= 3.12 it is possible to abbreviate cases where the field
being matched and the variable being bound have the same name, so now
you can just write:
let f { field } =
(* ... use field ... *)
(Similarly for a field prefixed by a Module name you can use
‘{ Module.field }’ instead of ‘{ Module.field = field }’).
This style is widely used inside the OCaml compiler sources, and is
briefer than the long form, so it makes sense to use it. Furthermore
there was one place in virt-dib where we are already using this new
style, so the old code did not compile on OCaml < 3.12.
See also:
https://forge.ocamlcore.org/docman/view.php/77/112/leroy-cug2010.pdf
This directory which previously contained random modules and functions
now has an official purpose: to be the place for any OCaml utility
needed by the OCaml virt tools.
This is just code movement, I didn't (yet) rename or move any of the
modules.
If configured with --without-ocaml, the build might fail because the
fix added in df5bd5741b was not active.
According to the automake documentation, it should be enough to set
BUILT_SOURCES.
The previously selected 'atomic' recipe resulted in 2GB swap in a 6GB
volume. Also, we don't really need the boot partition, so just create a
partition using the whole disk space.
These are generated in many different ways in the various
subdirectories, and sometimes not generated correctly. Introduce a
script to do this in one place, and hopefully correctly.
This is mostly simple refactoring, but I got rid of a couple of
things:
(1) The ‘make depend’ rule doesn't appear to be needed. automake (or
make?) seems to rebuild the ‘.depend’ file automatically just because
it is included.
(2) I got rid of the hairy path rewriting sed expression. Possibly
that is needed for srcdir != builddir.
The C function mkdtemp(3) requires that the string ends with 6 'X'
characters, so appending a non-empty suffix causes the function to
raise EINVAL.
Luckily we only ever called this function with the last parameter "".
