Merge branch 'decoupled' into dev

bopytex/__init__.py

@@ -1,10 +1,6 @@
 #!/usr/bin/env python
 # encoding: utf-8
 
-
-#from .bopytex import subject_metadatas, crazy_feed, pdfjoin
-
-
 # -----------------------------
 # Reglages pour 'vim'
 # vim:set autoindent expandtab tabstop=4 shiftwidth=4:

bopytex/bopytex.py

@@ -1,237 +1,2 @@
-#!/usr/bin/env python
-# encoding: utf-8
-
-"""
-Producing then compiling templates
-"""
-
-import csv
-import os
-import logging
-
-from pathlib import Path
-import pytex
-from mapytex import render, Expression, random, stat
-import bopytex.filters as filters
-import random as rd
-
-formatter = logging.Formatter("%(name)s :: %(levelname)s :: %(message)s")
-steam_handler = logging.StreamHandler()
-steam_handler.setLevel(logging.DEBUG)
-steam_handler.setFormatter(formatter)
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.DEBUG)
-logger.addHandler(steam_handler)
-
-
-def setup():
-    render.set_render("tex")
-    #logger.debug(f"Render for Expression is {Expression.RENDER}")
-    mapytex_tools = {
-        "Expression": Expression,
-        "rdm": random,
-        "stat": stat,
-        "random": rd,
-    }
-    pytex.update_export_dict(mapytex_tools)
-
-    pytex.add_filter("calculus", filters.do_calculus)
-
-
-def get_working_dir(options):
-    """ Get the working directory """
-    if options["working_dir"]:
-        working_dir = Path(options["working_dir"])
-    else:
-        try:
-            template = Path(options["template"])
-        except TypeError:
-            raise ValueError(
-                "Need to set the working directory \
-                    or to give a template"
-            )
-        else:
-            working_dir = template.parent
-    logger.debug(f"The output directory will be {working_dir}")
-    return working_dir
-
-
-def activate_printanswers(
-    texfile, noans=r"solution/print = false", ans=r"solution/print = true"
-):
-    """ Activate printanswers mod in texfile """
-    output_fname = "corr_" + texfile
-    with open(texfile, "r") as input_f:
-        with open(output_fname, "w") as output_f:
-            for line in input_f.readlines():
-                output_f.write(line.replace(noans, ans))
-    return output_fname
-
-
-def deactivate_printanswers(corr_fname):
-    """ Activate printanswers mod in texfile """
-    Path(corr_fname).remove()
-
-
-def pdfjoin(pdf_files, destname, working_dir=".", rm_pdfs=1):
-    """TODO: Docstring for pdfjoin.
-
-    :param pdf_files: list of pdf files to join
-    :param destname: name for joined pdf
-    :param working_dir: the working directory
-    :param rm_pdfs: Remove pdf_files after joining them
-    :returns: TODO
-
-    """
-    joined_pdfs = Path(working_dir) / Path(destname)
-    pdf_files_str = " ".join(pdf_files)
-    pdfjam = f"pdfjam {pdf_files_str} -o {joined_pdfs}"
-    logger.debug(f"Run {pdfjam}")
-    logger.info("Joining pdf files")
-    os.system(pdfjam)
-    if rm_pdfs:
-        logger.info(f"Remove {pdf_files_str}")
-        os.system(f"rm {pdf_files_str}")
-
-
-def extract_student_csv(csv_filename):
-    """ Extract student list from csv_filename """
-    with open(csv_filename, "r") as csvfile:
-        reader = csv.DictReader(csvfile)
-        return [r for r in reader]
-
-
-def subject_metadatas(options):
-    """ Return metadata on subject to produce
-
-    if csv is given it will based on is
-    otherwise it will be based on quantity
-
-    :example:
-    >>> subject_metadata(10)
-    """
-    if options["students_csv"]:
-        metadatas = []
-        for (i, s) in enumerate(extract_student_csv(options["students_csv"])):
-            d = {"num": f"{i+1:02d}"}
-            d.update(s)
-            metadatas.append(d)
-    elif options["number_subjects"] > 0:
-        metadatas = [{"num": f"{i+1:02d}"} for i in range(options["number_subjects"])]
-    else:
-        raise ValueError("Need metacsv or quantity to build subject metadata")
-
-    for meta in metadatas:
-        meta.update(
-            {
-                "template": str(Path(options["template"]).name),
-                "texfile": str(Path(options["template"]).name).replace(
-                    "tpl", meta["num"]
-                ),
-                "directory": str(Path(options["template"]).parent),
-            }
-        )
-
-    return metadatas
-
-
-def feed(*args, **kwrds):
-    """ Nice and smooth pytex feed """
-    pytex.feed(*args, **kwrds)
-
-
-def crazy_feed(*args, **kwrds):
-    """ Crazy mod for pytex feed """
-    while True:
-        try:
-            pytex.feed(*args, **kwrds)
-        except:
-            logger.debug(f"Crazy feed is working hard...! {args} {kwrds}")
-        else:
-            break
-
-
-def clean(directory):
-    pytex.clean(directory)
-
-
-def texcompile(filename):
-    logger.debug(f"Start compiling {filename}")
-    pytex.pdflatex(Path(filename))
-    logger.debug(f"End compiling")
-
-
-def produce_and_compile(options):
-    """ Produce and compile subjects
-    """
-    logger.debug(f"CI parser gets {options}")
-
-    template = Path(options["template"]).name
-    directory = Path(options["template"]).parent
-    metadatas = subject_metadatas(options)
-    logger.debug(f"Metadata {metadatas}")
-
-    setup()
-
-    for meta in metadatas:
-        logger.debug(f"Feeding template toward {meta['texfile']}")
-        if options["crazy"]:
-            crazy_feed(
-                template=Path(meta["directory"]) / meta["template"],
-                data=meta,
-                output=meta["texfile"],
-                force=1,
-            )
-        else:
-            feed(
-                template=Path(meta["directory"]) / meta["template"],
-                data=meta,
-                output=meta["texfile"],
-                force=1,
-            )
-        assert(Path(meta["texfile"]).exists())
-        logger.debug(f"{meta['texfile']} fed")
-
-        if options["corr"]:
-            logger.debug(f"Building correction for {meta['texfile']}")
-            meta.update({
-                "corr_texfile": activate_printanswers(meta["texfile"]),
-            })
-
-        if not options["no_compile"]:
-            logger.debug("Compiling")
-            for prefix in ["", "corr_"]:
-                key = prefix + "texfile"
-                try:
-                    meta[key]
-                except KeyError:
-                    pass
-                else:
-                    texcompile(meta[key])
-                    meta.update({
-                        prefix+'pdffile': meta[key].replace('tex', 'pdf')
-                    })
-
-    if not options["no_join"]:
-        for prefix in ["", "corr_"]:
-            key = prefix + "pdffile"
-            try:
-                pdfs = [m[key] for m in metadatas]
-            except KeyError:
-                pass
-            else:
-                pdfjoin(
-                    pdfs,
-                    template.replace("tpl", prefix+"all").replace(".tex", ".pdf"),
-                    directory,
-                    rm_pdfs=1,
-                )
-
-    if not options["dirty"]:
-        clean(directory)
-
-
-# -----------------------------
-# Reglages pour 'vim'
-# vim:set autoindent expandtab tabstop=4 shiftwidth=4:
-# cursor: 16 del
+import bopytex.default_config as DEFAULT
+from bopytex.service import orcherstrator

bopytex/default_config.py

@@ -0,0 +1,26 @@
+from bopytex.planner.generate_compile_join_planner import planner
+from bopytex.worker import Dispatcher
+from bopytex.worker.activate_corr import activate_corr
+from bopytex.worker.clean import clean
+from bopytex.worker.compile import pdflatex
+from bopytex.worker.generate import generate
+from bopytex.worker.join_pdf import pdfjam
+from bopytex.jinja2_env.texenv import texenv
+
+jinja2 = {
+    "environment": texenv
+}
+
+dispatcher = Dispatcher({
+    "GENERATE": generate,
+    "COMPILE": pdflatex,
+    "JOIN": pdfjam,
+    "CLEAN": clean,
+    "ACTIVATE_CORR": activate_corr,
+    })
+
+latex = {
+        "solution": r"solution/print = true",
+        "no_solution": r"solution/print = false",
+        }
+

bopytex/filters.py

@@ -1,33 +0,0 @@
-#!/usr/bin/env python
-# encoding: utf-8
-
-"""
-Custom filter for Bopytex
-"""
-
-__all__ = ["do_calculus"]
-
-def do_calculus(steps, name="A", sep="=", end="", joining=" \\\\ \n"):
-    """Display properly the calculus
-
-    Generate this form string:
-    "name & sep & a_step end joining"
-
-    :param steps: list of steps
-    :returns: latex string ready to be endbeded
-
-
-    """
-
-    ans = joining.join([
-        name + " & "
-        + sep + " & "
-        + str(s) + end for s in steps
-        ])
-    return ans
-
-
-# -----------------------------
-# Reglages pour 'vim'
-# vim:set autoindent expandtab tabstop=4 shiftwidth=4:
-# cursor: 16 del

bopytex/jinja2_env/texenv.py

@@ -0,0 +1,31 @@
+#!/usr/bin/env python
+# encoding: utf-8
+
+import jinja2
+
+__all__ = [
+    "texenv",
+]
+
+# Definition of jinja syntax for latex
+texenv = jinja2.Environment(
+    block_start_string=r"\Block{",
+    block_end_string="}",
+    variable_start_string=r"\Var{",
+    variable_end_string="}",
+    comment_start_string=r"\#{",
+    comment_end_string="}",
+    line_statement_prefix="%-",
+    line_comment_prefix="%#",
+    loader=jinja2.ChoiceLoader(
+        [
+            jinja2.FileSystemLoader(["./"]),
+        ]
+    ),
+    extensions=["jinja2.ext.do"],
+)
+
+# -----------------------------
+# Reglages pour 'vim'
+# vim:set autoindent expandtab tabstop=4 shiftwidth=4:
+# cursor: 16 del

bopytex/lib/pythagore.py

@@ -1,38 +0,0 @@
-#!/usr/bin/env python
-# encoding: utf-8
-
-from random import randint
-
-def pythagore_triplet(v_min = 1, v_max = 10):
-    """Random pythagore triplet generator
-
-    :param v_min: minimum in randint
-    :param v_max: max in randint
-    :returns: (a,b,c) such that a^2 + b^2 = c^2
-
-    """
-    u = randint(v_min,v_max)
-    v = randint(v_min,v_max)
-    while v == u:
-        v = randint(v_min,v_max)
-
-    u, v = max(u,v), min(u,v)
-
-    return (u**2-v**2 , 2*u*v, u**2 + v**2)
-
-if __name__ == '__main__':
-    print(pythagore_triplet())
-
-    for j in range(1,10):
-        for i in range(j,10):
-            print((i**2-j**2 , 2*i*j, i**2 + j**2))
-
-
-
-
-
-
-# -----------------------------
-# Reglages pour 'vim'
-# vim:set autoindent expandtab tabstop=4 shiftwidth=4:
-# cursor: 16 del 

bopytex/macros/poly2Deg.tex

@@ -1,33 +0,0 @@
-\Block{macro solveEquation(P)}
-
-    On commence par calculer le discriminant de $P(x) = \Var{P}$.
-    \begin{eqnarray*}
-        \Delta & = & b^2-4ac \\
-        \Var{P.delta.explain()|calculus(name="\\Delta")}
-    \end{eqnarray*}
-
-    \Block{if P.delta > 0}
-    comme $\Delta = \Var{P.delta} > 0$ donc $P$ a deux racines
-
-    \begin{eqnarray*}
-        x_1 & = & \frac{-b - \sqrt{\Delta}}{2a} =  \frac{\Var{-P.b} - \sqrt{\Var{P.delta}}}{2 \times \Var{P.a}} = \Var{P.roots()[0] } \\
-        x_2 & = & \frac{-b + \sqrt{\Delta}}{2a} =  \frac{\Var{-P.b} + \sqrt{\Var{P.delta}}}{2 \times \Var{P.a}} = \Var{P.roots()[1] }
-    \end{eqnarray*}
-
-    Les solutions de l'équation $\Var{P} = 0$ sont donc $\mathcal{S} = \left\{ \Var{P.roots()[0]}; \Var{P.roots()[1]} \right\}$
-
-    \Block{elif P.delta == 0}
-    Comme $\Delta = 0$ donc $P$ a une racine
-
-    \begin{eqnarray*}
-        x_1 = \frac{-b}{2a} = \frac{-\Var{P.b}}{2\times \Var{P.a}} = \Var{P.roots()[0]} \\
-    \end{eqnarray*}
-
-    La solution de $\Var{P} = 0$ est donc $\mathcal{S} = \left\{ \Var{P.roots()[0]}\right\}$
-
-    \Block{else}
-    Alors $\Delta = \Var{P.delta} < 0$ donc $P$ n'a pas de racine donc l'équation $\Var{P} = 0$ n'a pas de solution.
-
-    \Block{endif}
-
-\Block{endmacro}

bopytex/message.py

@@ -0,0 +1,36 @@
+class Message():
+    def __init__(self, status, out, err):
+        self._status = status
+        self._out = out
+        self._err = err
+
+    @property
+    def status(self):
+        return self._status
+
+    @property
+    def out(self):
+        return self._out
+
+    @property
+    def err(self):
+        return self._err
+
+    def __repr__(self):
+        return f"Message(status={self.status}, out={self.out}, err={self.err})"
+
+class SubprocessMessage(Message):
+    def __init__(self, process):
+        self._process = process
+
+    @property
+    def status(self):
+        return self._process.wait()
+
+    @property
+    def out(self):
+        return self._process.stdout
+
+    @property
+    def err(self):
+        return self._process.stderr

bopytex/planner/activate_corr_compile_join_planner.py

@@ -0,0 +1,58 @@
+from bopytex.tasks import Task, activate_corr_on, compile_pdf, join_pdfs
+import bopytex.planner.naming as naming
+import os
+
+
+def list_files(dir=".", accept=lambda _: True, reject=lambda _: False):
+    files = []
+    for file in os.listdir(dir):
+        if accept(file) and not reject(file):
+            files.append(file)
+    return files
+
+
+def planner(options: dict) -> list[Task]:
+    sources = list_files(
+        accept=lambda x: x.endswith(".tex"),
+        reject=lambda x: x.startswith("tpl_"),
+    )
+    options["sources"] = sources
+    return tasks_builder(options)
+
+
+def tasks_builder(
+    options: dict,
+) -> list[Task]:
+    opt = {
+        "no_join": False,
+        "no_pdf": False,
+    }
+    opt.update(options)
+
+    try:
+        sources = opt["sources"]
+        no_join = opt["no_join"]
+        no_pdf = opt["no_pdf"]
+    except KeyError:
+        raise PlannerMissingOption("An option is missing")
+
+    tasks = []
+    corr_pdfs = []
+
+    for source in sources:
+        corr_source = naming.corr(source)
+        tasks.append(activate_corr_on(source, opt, corr_source))
+
+        if not no_pdf:
+            corr_pdf = naming.source2pdf(corr_source)
+            tasks.append(compile_pdf(corr_source, corr_pdf))
+            corr_pdfs.append(corr_pdf)
+
+    if not no_join:
+        joined = "joined.pdf"
+
+        if corr_pdfs:
+            corr_joined = naming.corr(joined)
+            tasks.append(join_pdfs(corr_pdfs, corr_joined))
+
+    return tasks

bopytex/planner/exceptions.py

@@ -0,0 +1,2 @@
+class PlannerMissingOption(Exception):
+    pass

bopytex/planner/fake_planner.py

@@ -0,0 +1,9 @@
+from bopytex.tasks import Task
+
+
+def simple(options: dict) -> list[Task]:
+    """Simple planner with options['quantity'] tasks and no dependencies"""
+    return [
+        Task("DO", args={"number": i}, deps=[], output=f"{i}")
+        for i in range(options["quantity"])
+    ]

bopytex/planner/generate_compile_join_planner.py

@@ -0,0 +1,103 @@
+from bopytex.tasks import Task, activate_corr_on, compile_pdf, generate, join_pdfs
+import bopytex.planner.naming as naming
+from bopytex.planner.exceptions import PlannerMissingOption
+import csv
+
+
+def build_subject_list_from_infos(infos: list[dict]) -> list[dict]:
+    subjects = []
+    digit = len(str(len(infos)))
+    for i, infos in enumerate(infos):
+        subjects.append({"number": str(i + 1).zfill(digit), **infos})
+    return subjects
+
+
+def build_subject_list_from_qty(qty: int) -> list[dict]:
+    subjects = []
+    digit = len(str(qty))
+    for i in range(qty):
+        subjects.append({"number": str(i + 1).zfill(digit)})
+    return subjects
+
+
+def planner(options: dict) -> list[Task]:
+    try:
+        students_csv = options["students_csv"]
+        assert options["students_csv"] != ""
+
+    except (KeyError, AssertionError):
+        try:
+            quantity_subjects = options["quantity_subjects"]
+            assert options["quantity_subjects"] != 0
+        except (KeyError, AssertionError):
+            raise PlannerMissingOption("students_csv or quantity_subjects is required")
+        else:
+            options["subjects"] = build_subject_list_from_qty(qty=quantity_subjects)
+
+    else:
+        with open(students_csv, "r") as csv_file:
+            reader = csv.DictReader(csv_file)
+            infos = [r for r in reader]
+        options["subjects"] = build_subject_list_from_infos(infos)
+
+    return tasks_builder(options)
+
+
+def tasks_builder(
+    options: dict,
+) -> list[Task]:
+
+    opt = {
+        "corr": False,
+        "no_join": False,
+        "no_pdf": False,
+    }
+    opt.update(options)
+
+    try:
+        template = opt["template"]
+        subjects = opt["subjects"]
+        corr = opt["corr"]
+        no_join = opt["no_join"]
+        no_pdf = opt["no_pdf"]
+    except KeyError:
+        raise PlannerMissingOption("An option is missing")
+
+    tasks = []
+
+    pdfs = []
+    corr_pdfs = []
+
+    for subject in subjects:
+        source = naming.template2source(template, subject)
+        args = {
+                "subject": subject, 
+                "options": options
+                }
+
+        tasks.append(generate(template, args, source))
+
+        if not no_pdf:
+            pdf = naming.source2pdf(source)
+            tasks.append(compile_pdf(source, pdf))
+            pdfs.append(pdf)
+
+        if corr:
+            corr_source = naming.corr(source)
+            tasks.append(activate_corr_on(source, opt, corr_source))
+
+            if not no_pdf:
+                corr_pdf = naming.source2pdf(corr_source)
+                tasks.append(compile_pdf(corr_source, corr_pdf))
+                corr_pdfs.append(corr_pdf)
+
+    if not no_join:
+        joined = naming.join(template)
+        if pdfs:
+            tasks.append(join_pdfs(pdfs, joined))
+
+        if corr_pdfs:
+            corr_joined = naming.corr(joined)
+            tasks.append(join_pdfs(corr_pdfs, corr_joined))
+
+    return tasks

bopytex/planner/naming.py

@@ -0,0 +1,15 @@
+def template2source(template: str, metadatas: dict):
+    return metadatas["number"] + template[3:]
+
+
+def corr(source):
+    return "corr_" + source
+
+
+def source2pdf(source):
+    return source[:-4] + ".pdf"
+
+
+def join(template):
+    return source2pdf("joined" + template[3:])
+

bopytex/scheduler.py

@@ -0,0 +1,78 @@
+""" Scheduler for action to make """
+
+
+from bopytex.tasks import Task
+from bopytex.worker import Dispatcher
+
+
+class Scheduler:
+    """Scheduler is responsible of getting tasks (the tasks) and yield those that can be done"""
+
+    def __init__(self, dispatcher:Dispatcher, output_done: list[str] = None):
+        self._dispatcher = dispatcher
+
+        if output_done is None:
+            self._output_done = []
+        else:
+            self._output_done = output_done
+
+        self._tasks = []
+        self._failed_tasks = []
+
+    @property
+    def tasks(self) -> list[Task]:
+        """List all the tasks todo"""
+        return self._tasks
+
+    @property
+    def doable_tasks(self) -> list[Task]:
+        """List all doable tasks"""
+        return [
+            task
+            for task in self.tasks
+            if not task.deps or all([d in self.output_done for d in task.deps])
+        ]
+
+    @property
+    def all_deps(self) -> set[str]:
+        """List dependencies of all tasks"""
+        return {d for task in self.tasks for d in task.deps}
+
+    @property
+    def all_output(self) -> set[str]:
+        """List ouput of all tasks"""
+        return {task.output for task in self.tasks}
+
+    @property
+    def output_done(self) -> list[str]:
+        return self._output_done
+
+    @property
+    def failed_tasks(self) -> list[Task]:
+        return self._failed_tasks
+
+    def append(self, tasks: list[Task]):
+        self._tasks += tasks
+
+    def is_finishable(self):
+        return self.all_deps.issubset(self.all_output)
+
+    def next_task(self):
+        try:
+            task = self.doable_tasks[0]
+        except IndexError:
+            raise StopIteration
+
+        self._tasks.remove(task)
+        message = self._dispatcher(task)
+        if message.status == 0:
+            self._output_done.append(task.output)
+        else:
+            self._failed_tasks.append(task)
+
+        return message
+
+    def backlog(self):
+        """ Yield tasks sorted according to dependencies """
+        while self.doable_tasks:
+            yield self.next_task()

bopytex/script.py

@@ -3,20 +3,30 @@
 
 
 import click
-from .bopytex import produce_and_compile
+import logging
+
+from bopytex.service import main
+
+formatter = logging.Formatter("%(name)s :: %(levelname)s :: %(message)s")
+steam_handler = logging.StreamHandler()
+steam_handler.setLevel(logging.DEBUG)
+steam_handler.setFormatter(formatter)
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.DEBUG)
+logger.addHandler(steam_handler)
+
 
 @click.command()
 @click.argument(
     "template",
     type=click.Path(exists=True),
     nargs=1,
-    # help="File with the template. The name should have the following form tpl_... .",
 )
 @click.option(
     "-w",
     "--working-dir",
+    default=".",
     type=click.Path(exists=True),
-    help="Where fed templates and compiled files will be placed",
 )
 @click.option(
     "-s",
@@ -25,9 +35,6 @@ from .bopytex import produce_and_compile
     default="",
     help="CSV containing list of students names",
 )
-@click.option(
-    "-d", "--dirty", is_flag=True, default=False, help="Do not clean after compilation",
-)
 @click.option(
     "-n",
     "--no-compile",
@@ -36,11 +43,18 @@ from .bopytex import produce_and_compile
     help="Do not compile source code",
 )
 @click.option(
-    "-N",
-    "--number_subjects",
+    "-d",
+    "--dirty",
+    is_flag=True,
+    default=False,
+    help="Do not clean after compilation",
+)
+@click.option(
+    "-q",
+    "--quantity_subjects",
     type=int,
     default=1,
-    help="The number of subjects to make",
+    help="The quantity of subjects to make",
 )
 @click.option(
     "-j",
@@ -54,29 +68,31 @@ from .bopytex import produce_and_compile
     "--only-corr",
     is_flag=True,
     default=False,
-    help="Create and compile only correction from existing subjects",
+    help="Activate correction and compile only from existing subjects",
 )
 @click.option(
-    "-c",
+    "-C",
     "--corr",
     is_flag=True,
     default=False,
     help="Create and compile correction while making subjects",
 )
 @click.option(
-    "-C",
-    "--crazy",
-    is_flag=True,
-    default=False,
-    help="Crazy mode. Tries and tries again until template feeding success!",
+    "-c",
+    "--configfile",
+    type=str,
+    default="bopyptex_config.py",
+    help="Config file path",
 )
 def new(**options):
-    """ Bopytex
-
-    Feed the template (tpl_...) and then compile it with latex.
-
-    """
-    produce_and_compile(options)
+    for message in main(**options):
+        try:
+            assert message.status == 0
+        except AssertionError:
+            logger.warning(message)
+            break
+        else:
+            logger.info(message.out)
 
 
 if __name__ == "__main__":

bopytex/service.py

@@ -0,0 +1,99 @@
+#!/usr/bin/env python
+# encoding: utf-8
+
+"""
+Producing then compiling templates
+"""
+
+import importlib.util
+import os
+from pathlib import Path
+from bopytex.scheduler import Scheduler
+from bopytex import default_config
+
+
+def orcherstrator(
+    options: dict,
+    planner,
+    dispatcher,
+):
+    tasks = planner(options)
+
+    scheduler = Scheduler(dispatcher, [options["template"]])
+    scheduler.append(tasks)
+
+    for message in scheduler.backlog():
+        yield message
+
+
+def load_module(modulefile: str):
+    spec = importlib.util.spec_from_file_location("module.name", modulefile)
+    module = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(module)
+    return module
+
+
+def clean_vars_keys(
+    vars: dict,
+    keys: list[str] = [
+        "__name__",
+        "__doc__",
+        "__package__",
+        "__loader__",
+        "__spec__",
+        "__file__",
+        "__cached__",
+        "__builtins__",
+    ],
+) -> dict:
+    new_dict = vars.copy()
+    for k in keys:
+        del new_dict[k]
+    return new_dict
+
+
+def config_from_file(filename: str) -> dict:
+    if Path(filename).exists():
+        local_config = vars(load_module(filename))
+        return clean_vars_keys(local_config)
+    else:
+        return {}
+
+
+def build_config(options: dict) -> dict:
+    """Look for options["configfile"] to load it with default_config and options"""
+    config = clean_vars_keys(vars(default_config))
+
+    configfile = ""
+    try:
+        configfile = options["configfile"]
+    except KeyError:
+        pass
+    try:
+        configfile = os.environ["BOPYTEXCONFIG"]
+    except KeyError:
+        pass
+    if configfile:
+        local_config = config_from_file(configfile)
+        config.update(local_config)
+
+    config.update(options)
+
+    return config
+
+
+def main(**options):
+
+    config = build_config(options)
+
+    orcherstre = orcherstrator(
+        config, planner=default_config.planner, dispatcher=default_config.dispatcher
+    )
+    for message in orcherstre:
+        yield message
+
+
+# -----------------------------
+# Reglages pour 'vim'
+# vim:set autoindent expandtab tabstop=4 shiftwidth=4:
+# cursor: 16 del

bopytex/tasks.py

@@ -0,0 +1,64 @@
+""" Produce tasks to do
+
+It essentially place things at the right place.
+
+"""
+from dataclasses import dataclass
+
+
+@dataclass
+class Task:
+    action: str
+    args: dict
+    deps: list
+    output: str
+
+
+def generate(template: str, meta: dict, output: str):
+    """Create a task to generate a subject"""
+    return Task(
+        action="GENERATE",
+        args=meta,
+        deps=[template],
+        output=output,
+    )
+
+
+def activate_corr_on(src: str, meta:dict, output: str):
+    """Create a task to activate correction for src"""
+    return Task(
+        action="ACTIVATE_CORR",
+        args=meta,
+        deps=[src],
+        output=output,
+    )
+
+
+def compile_pdf(src: str, output: str):
+    """Create a task to compile src"""
+    return Task(
+        action="COMPILE",
+        args={},
+        deps=[src],
+        output=output,
+    )
+
+
+def join_pdfs(pdfs: list, output: str):
+    """Create task to join pdf together"""
+    return Task(
+        action="JOIN",
+        args={},
+        deps=pdfs,
+        output=output,
+    )
+
+
+def clean(files: list):
+    """Create task to clean files"""
+    return Task(
+        action="CLEAN",
+        args={},
+        deps=files,
+        output=None,
+    )

bopytex/worker/__init__.py

@@ -0,0 +1,18 @@
+class ActionNotFound(Exception):
+    pass
+
+
+class Dispatcher:
+    def __init__(self, actions: list):
+        self._actions = actions
+
+    def __call__(self, task):
+        try:
+            choosen_action = self._actions[task.action]
+        except KeyError:
+            raise ActionNotFound(
+                f"The action {task.action} is not in {self._actions.keys()}"
+            )
+
+        return choosen_action(args=task.args, deps=task.deps, output=task.output)
+

bopytex/worker/activate_corr.py

@@ -0,0 +1,14 @@
+from bopytex.message import Message
+
+
+def activate_corr(args, deps, output):
+    no_solution = args["latex"]["no_solution"]
+    solution = args["latex"]["solution"]
+
+    with open(deps[0], "r") as input_f:
+        with open(output, "w") as output_f:
+            for line in input_f.readlines():
+                output_f.write(line.replace(no_solution, solution))
+
+    return Message(0, [f"ACTIVATE CORR - {deps[0]} to {output}"], [])
+

bopytex/worker/clean.py

@@ -0,0 +1,2 @@
+def clean(args: dict, deps, output):
+    pass

bopytex/worker/compile.py

@@ -0,0 +1,25 @@
+import subprocess
+
+from bopytex.message import Message
+from ..message import SubprocessMessage
+
+
+def curstomtex(command: str, options: str):
+    def func(args: dict, deps, output) -> Message:
+        compile_process = subprocess.Popen(
+            [command, options, deps[0]],
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+            universal_newlines=True,
+        )
+        return Message(
+            compile_process.wait(),
+            list(compile_process.stdout),
+            list(compile_process.stderr),
+        )
+
+    return func
+
+
+latexmk = curstomtex("latexmk", "-f")
+pdflatex = curstomtex("pdflatex", "--interaction=nonstopmode")

bopytex/worker/generate.py

@@ -0,0 +1,32 @@
+from jinja2.environment import Template
+
+from bopytex.message import Message
+
+
+def generate(args, deps, output):
+    env = args["options"]["jinja2"]["environment"]
+    template = env.get_template(deps[0])
+
+    variables = {
+            "options":args["options"],
+            "subject":args["subject"],
+            }
+
+    try:
+        args["options"]["direct_access"]
+    except KeyError:
+        pass
+    else:
+        for (k,v) in args["options"]["direct_access"].items():
+            if k not in ["options", "subject"]:
+                variables[k] = v
+
+    try:
+        with open(output, "w") as out:
+            fed = template.render(variables)
+            out.write(fed)
+
+        return Message(0, [f"GENERATE - {deps[0]} to {output}"], [])
+
+    except Exception as e:
+        return Message(1, [], [e])

bopytex/worker/join_pdf.py

@@ -0,0 +1,33 @@
+import subprocess
+
+from bopytex.message import Message, SubprocessMessage
+
+
+def pdfjam(args: dict, deps, output):
+    joining_process = subprocess.Popen(
+        ["pdfjam"] + deps + ["-o", output],
+        stdout=subprocess.PIPE,
+        stderr=subprocess.PIPE,
+        universal_newlines=True,
+    )
+    return Message(
+        joining_process.wait(),
+        list(joining_process.stdout),
+        list(joining_process.stderr),
+    )
+
+
+def gs(args: dict, deps, output):
+    """ Not working. The command works in terminal but not here """
+    joining_process = subprocess.Popen(
+        ["gs", f"-dBATCH -dNOPAUSE -q -sDEVICE=pdfwrite -sOutputFile={output}"]
+        + deps,
+        stdout=subprocess.PIPE,
+        stderr=subprocess.PIPE,
+        universal_newlines=True,
+    )
+    return Message(
+        joining_process.wait(),
+        list(joining_process.stdout),
+        list(joining_process.stderr),
+    )