Towards managing mapping for file imports.

This commit is contained in:
James Cole
2018-05-06 16:19:29 +02:00
parent 9e3c5fd984
commit a3cbdadb39
11 changed files with 868 additions and 84 deletions

View File

@@ -30,6 +30,15 @@ use Illuminate\Support\MessageBag;
*/
interface ConfigurationInterface
{
/**
* Store data associated with current stage.
*
* @param array $data
*
* @return MessageBag
*/
public function configureJob(array $data): MessageBag;
/**
* Get the data necessary to show the configuration screen.
*
@@ -39,17 +48,6 @@ interface ConfigurationInterface
/**
* @param ImportJob $job
*
* @return ConfigurationInterface
*/
public function setJob(ImportJob $job);
/**
* Store data associated with current stage.
*
* @param array $data
*
* @return MessageBag
*/
public function configureJob(array $data): MessageBag;
public function setJob(ImportJob $job): void;
}

View File

@@ -0,0 +1,296 @@
<?php
/**
* ConfigureMappingHandler.php
* Copyright (c) 2018 thegrumpydictator@gmail.com
*
* This file is part of Firefly III.
*
* Firefly III is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Firefly III is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with Firefly III. If not, see <http://www.gnu.org/licenses/>.
*/
declare(strict_types=1);
namespace FireflyIII\Support\Import\Configuration\File;
use FireflyIII\Exceptions\FireflyException;
use FireflyIII\Helpers\Attachments\AttachmentHelperInterface;
use FireflyIII\Import\Mapper\MapperInterface;
use FireflyIII\Models\Attachment;
use FireflyIII\Models\ImportJob;
use FireflyIII\Repositories\ImportJob\ImportJobRepositoryInterface;
use Illuminate\Support\Collection;
use Illuminate\Support\MessageBag;
use League\Csv\Exception;
use League\Csv\Reader;
use Log;
/**
* Class ConfigureMappingHandler
*/
class ConfigureMappingHandler implements ConfigurationInterface
{
/** @var AttachmentHelperInterface */
private $attachments;
/** @var array */
private $columnConfig;
/** @var ImportJob */
private $importJob;
/** @var ImportJobRepositoryInterface */
private $repository;
/**
* Store data associated with current stage.
*
* @param array $data
*
* @return MessageBag
*/
public function configureJob(array $data): MessageBag
{
return new MessageBag;
}
/**
* Get the data necessary to show the configuration screen.
*
* @return array
* @throws FireflyException
*/
public function getNextData(): array
{
$config = $this->importJob->configuration;
$columnConfig = $this->doColumnConfig($config);
// in order to actually map we also need to read the FULL file.
try {
$reader = $this->getReader();
} catch (Exception $e) {
Log::error($e->getMessage());
throw new FireflyException('Cannot get reader: ' . $e->getMessage());
}
//
// if ($config['has-headers']) {
// $offset = 1;
// }
// $stmt = (new Statement)->offset($offset);
// $results = $stmt->process($reader);
// $this->validSpecifics = array_keys(config('csv.import_specifics'));
// $indexes = array_keys($this->data);
// $rowIndex = 0;
// foreach ($results as $rowIndex => $row) {
// $row = $this->runSpecifics($row);
//
// //do something here
// foreach ($indexes as $index) { // this is simply 1, 2, 3, etc.
// if (!isset($row[$index])) {
// // don't really know how to handle this. Just skip, for now.
// continue;
// }
// $value = trim($row[$index]);
// if (\strlen($value) > 0) {
// // we can do some preprocessing here,
// // which is exclusively to fix the tags:
// if (null !== $this->data[$index]['preProcessMap'] && \strlen($this->data[$index]['preProcessMap']) > 0) {
// /** @var PreProcessorInterface $preProcessor */
// $preProcessor = app($this->data[$index]['preProcessMap']);
// $result = $preProcessor->run($value);
// $this->data[$index]['values'] = array_merge($this->data[$index]['values'], $result);
//
// Log::debug($rowIndex . ':' . $index . 'Value before preprocessor', ['value' => $value]);
// Log::debug($rowIndex . ':' . $index . 'Value after preprocessor', ['value-new' => $result]);
// Log::debug($rowIndex . ':' . $index . 'Value after joining', ['value-complete' => $this->data[$index]['values']]);
//
// continue;
// }
//
// $this->data[$index]['values'][] = $value;
// }
// }
// }
// $setIndexes = array_keys($this->data);
// foreach ($setIndexes as $index) {
// $this->data[$index]['values'] = array_unique($this->data[$index]['values']);
// asort($this->data[$index]['values']);
// // if the count of this array is zero, there is nothing to map.
// if (\count($this->data[$index]['values']) === 0) {
// unset($this->data[$index]);
// }
// }
// unset($setIndexes);
//
// // save number of rows, thus number of steps, in job:
// $steps = $rowIndex * 5;
// $extended = $this->job->extended_status;
// $extended['steps'] = $steps;
// $this->job->extended_status = $extended;
// $this->job->save();
//
// return $this->data;
// */
}
/**
* @param ImportJob $job
*/
public function setJob(ImportJob $job): void
{
$this->importJob = $job;
$this->repository = app(ImportJobRepositoryInterface::class);
$this->repository->setUser($job->user);
$this->attachments = app(AttachmentHelperInterface::class);
$this->columnConfig = [];
}
/**
* Create the "mapper" class that will eventually return the correct data for the user
* to map against. For example: a list of asset accounts. A list of budgets. A list of tags.
*
* @param string $column
*
* @return MapperInterface
* @throws FireflyException
*/
private function createMapper(string $column): MapperInterface
{
$mapperClass = config('csv.import_roles.' . $column . '.mapper');
$mapperName = sprintf('\\FireflyIII\\Import\Mapper\\%s', $mapperClass);
if (!class_exists($mapperName)) {
throw new FireflyException(sprintf('Class "%s" does not exist. Cannot map "%s"', $mapperName, $column));
}
return app($mapperName);
}
/**
* For each column in the configuration of the job, will:
* - validate the role.
* - validate if it can be used for mapping
* - if so, create an entry in $columnConfig
*
* @param array $config
*
* @return array the column configuration.
* @throws FireflyException
*/
private function doColumnConfig(array $config): array
{
/** @var array $requestMapping */
$requestMapping = $config['column-do-mapping'] ?? [];
$columnConfig = [];
/**
* @var int
* @var bool $mustBeMapped
*/
foreach ($requestMapping as $index => $requested) {
// sanitize column name, so we're sure it's valid.
$column = $this->sanitizeColumnName($config['column-roles'][$index] ?? '_ignore');
$doMapping = $this->doMapOfColumn($column, $requested);
if ($doMapping) {
// user want to map this column. And this is possible.
$columnConfig[$index] = [
'name' => $column,
'options' => $this->createMapper($column)->getMap(),
'preProcessMap' => $this->getPreProcessorName($column),
'values' => [],
];
}
}
return $columnConfig;
}
/**
* For each $name given, and if the user wants to map the column, will return
* true when the column can also be mapped.
*
* Unmappable columns will always return false.
* Mappable columns will return $requested.
*
* @param string $name
* @param bool $requested
*
* @return bool
*/
private function doMapOfColumn(string $name, bool $requested): bool
{
$canBeMapped = config('csv.import_roles.' . $name . '.mappable');
return $canBeMapped && $requested;
}
/**
* Will return the name of the pre-processor: a special class that will clean up any input that may be found
* in the users input (aka the file uploaded). Only two examples exist at this time: a space or comma separated
* list of tags.
*
* @param string $column
*
* @return string
*/
private function getPreProcessorName(string $column): string
{
$name = '';
$hasPreProcess = config(sprintf('csv.import_roles.%s.pre-process-map', $column));
$preProcessClass = config(sprintf('csv.import_roles.%s.pre-process-mapper', $column));
if (null !== $hasPreProcess && true === $hasPreProcess && null !== $preProcessClass) {
$name = sprintf('\\FireflyIII\\Import\\MapperPreProcess\\%s', $preProcessClass);
}
return $name;
}
/**
* Return an instance of a CSV file reader so content of the file can be read.
*
* @throws \League\Csv\Exception
*/
private function getReader(): Reader
{
$content = '';
/** @var Collection $collection */
$collection = $this->importJob->attachments;
/** @var Attachment $attachment */
foreach ($collection as $attachment) {
if ($attachment->filename === 'import_file') {
$content = $this->attachments->getAttachmentContent($attachment);
break;
}
}
$config = $this->repository->getConfiguration($this->importJob);
$reader = Reader::createFromString($content);
$reader->setDelimiter($config['delimiter']);
return $reader;
}
/**
* For each given column name, will return either the name (when it's a valid one)
* or return the _ignore column.
*
* @param string $name
*
* @return string
*/
private function sanitizeColumnName(string $name): string
{
/** @var array $validColumns */
$validColumns = array_keys(config('csv.import_roles'));
if (!\in_array($name, $validColumns, true)) {
$name = '_ignore';
}
return $name;
}
}

View File

@@ -0,0 +1,394 @@
<?php
/**
* ConfigureRolesHandler.php
* Copyright (c) 2018 thegrumpydictator@gmail.com
*
* This file is part of Firefly III.
*
* Firefly III is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Firefly III is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with Firefly III. If not, see <http://www.gnu.org/licenses/>.
*/
declare(strict_types=1);
namespace FireflyIII\Support\Import\Configuration\File;
use FireflyIII\Exceptions\FireflyException;
use FireflyIII\Helpers\Attachments\AttachmentHelperInterface;
use FireflyIII\Import\Specifics\SpecificInterface;
use FireflyIII\Models\Attachment;
use FireflyIII\Models\ImportJob;
use FireflyIII\Repositories\ImportJob\ImportJobRepositoryInterface;
use Illuminate\Support\Collection;
use Illuminate\Support\MessageBag;
use League\Csv\Exception;
use League\Csv\Reader;
use League\Csv\Statement;
use Log;
/**
* Class ConfigureRolesHandler
*/
class ConfigureRolesHandler implements ConfigurationInterface
{
/** @var AttachmentHelperInterface */
private $attachments;
/** @var array */
private $examples;
/** @var ImportJob */
private $importJob;
/** @var ImportJobRepositoryInterface */
private $repository;
/** @var int */
private $totalColumns;
/**
* Store data associated with current stage.
*
* @param array $data
*
* @return MessageBag
*/
public function configureJob(array $data): MessageBag
{
$config = $this->importJob->configuration;
$count = $config['column-count'];
for ($i = 0; $i < $count; ++$i) {
$role = $data['role'][$i] ?? '_ignore';
$mapping = (isset($data['map'][$i]) && $data['map'][$i] === '1');
$config['column-roles'][$i] = $role;
$config['column-do-mapping'][$i] = $mapping;
Log::debug(sprintf('Column %d has been given role %s (mapping: %s)', $i, $role, var_export($mapping, true)));
}
$config = $this->ignoreUnmappableColumns($config);
$messages = $this->configurationComplete($config);
if ($messages->count() === 0) {
$this->repository->setStage($this->importJob, 'ready_to_run');
if ($this->isMappingNecessary($config)) {
$this->repository->setStage($this->importJob, 'map');
}
$this->repository->setConfiguration($this->importJob, $config);
}
return $messages;
}
/**
* Get the data necessary to show the configuration screen.
*
* @return array
* @throws FireflyException
*/
public function getNextData(): array
{
try {
$reader = $this->getReader();
} catch (Exception $e) {
Log::error($e->getMessage());
throw new FireflyException($e->getMessage());
}
$headers = $this->getHeaders($reader);
// get example rows:
$this->getExamples($reader);
return [
'examples' => $this->examples,
'roles' => $this->getRoles(),
'total' => $this->totalColumns,
'headers' => $headers,
];
}
/**
* Set job and some start values.
*
* @param ImportJob $job
*/
public function setJob(ImportJob $job): void
{
$this->importJob = $job;
$this->repository = app(ImportJobRepositoryInterface::class);
$this->repository->setUser($job->user);
$this->attachments = app(AttachmentHelperInterface::class);
$this->totalColumns = 0;
$this->examples = [];
}
/**
* Verifies that the configuration of the job is actually complete, and valid.
*
* @param array $config
*
* @return MessageBag
*/
private function configurationComplete(array $config): MessageBag
{
$count = $config['column-count'];
$assigned = 0;
// check if data actually contains amount column (foreign amount does not count)
$hasAmount = false;
$hasForeignAmount = false;
$hasForeignCode = false;
for ($i = 0; $i < $count; ++$i) {
$role = $config['column-roles'][$i] ?? '_ignore';
if ('_ignore' !== $role) {
++$assigned;
}
if (\in_array($role, ['amount', 'amount_credit', 'amount_debit'])) {
$hasAmount = true;
}
if ($role === 'foreign-currency-code') {
$hasForeignCode = true;
}
if ($role === 'amount_foreign') {
$hasForeignAmount = true;
}
}
// all assigned and correct foreign info
if ($assigned > 0 && $hasAmount && ($hasForeignCode === $hasForeignAmount)) {
return new MessageBag;
}
if (0 === $assigned || !$hasAmount) {
$message = (string)trans('import.job_config_roles_rwarning');
$messages = new MessageBag();
$messages->add('error', $message);
return $messages;
}
// warn if has foreign amount but no currency code:
if ($hasForeignAmount && !$hasForeignCode) {
$message = (string)trans('import.job_config_roles_fa_warning');
$messages = new MessageBag();
$messages->add('error', $message);
return $messages;
}
return new MessageBag;
}
/**
* Extracts example data from a single row and store it in the class.
*
* @param array $row
*/
private function getExampleFromRow(array $row): void
{
foreach ($row as $index => $value) {
$value = trim($value);
if (\strlen($value) > 0) {
$this->examples[$index][] = $value;
}
}
}
/**
* Return a bunch of examples from the CSV file the user has uploaded.
*
* @param Reader $reader
*
* @throws FireflyException
*/
private function getExamples(Reader $reader): void
{
// configure example data:
$config = $this->importJob->configuration;
$limit = (int)config('csv.example_rows', 5);
$offset = isset($config['has-headers']) && $config['has-headers'] === true ? 1 : 0;
// make statement.
try {
$stmt = (new Statement)->limit($limit)->offset($offset);
} catch (Exception $e) {
Log::error($e->getMessage());
throw new FireflyException($e->getMessage());
}
// grab the records:
$records = $stmt->process($reader);
/** @var array $row */
foreach ($records as $row) {
$row = array_values($row);
$row = $this->processSpecifics($row);
$count = \count($row);
$this->totalColumns = $count > $this->totalColumns ? $count : $this->totalColumns;
$this->getExampleFromRow($row);
}
// save column count:
$this->saveColumCount();
$this->makeExamplesUnique();
}
/**
* Get the header row, if one is present.
*
* @param Reader $reader
*
* @return array
* @throws FireflyException
*/
private function getHeaders(Reader $reader): array
{
$headers = [];
$config = $this->importJob->configuration;
if ($config['has-headers']) {
try {
$stmt = (new Statement)->limit(1)->offset(0);
$records = $stmt->process($reader);
$headers = $records->fetchOne(0);
} catch (Exception $e) {
Log::error($e->getMessage());
throw new FireflyException($e->getMessage());
}
Log::debug('Detected file headers:', $headers);
}
return $headers;
}
/**
* Return an instance of a CSV file reader so content of the file can be read.
*
* @throws \League\Csv\Exception
*/
private function getReader(): Reader
{
$content = '';
/** @var Collection $collection */
$collection = $this->importJob->attachments;
/** @var Attachment $attachment */
foreach ($collection as $attachment) {
if ($attachment->filename === 'import_file') {
$content = $this->attachments->getAttachmentContent($attachment);
break;
}
}
$config = $this->repository->getConfiguration($this->importJob);
$reader = Reader::createFromString($content);
$reader->setDelimiter($config['delimiter']);
return $reader;
}
/**
* Returns all possible roles and translate their name. Then sort them.
*
* @return array
*/
private function getRoles(): array
{
$roles = [];
foreach (array_keys(config('csv.import_roles')) as $role) {
$roles[$role] = trans('import.column_' . $role);
}
asort($roles);
return $roles;
}
/**
* If the user has checked columns that cannot be mapped to any value, this function will
* uncheck them and return the configuration again.
*
* @param array $config
*
* @return array
*/
private function ignoreUnmappableColumns(array $config): array
{
$count = $config['column-count'];
for ($i = 0; $i < $count; ++$i) {
$role = $config['column-roles'][$i] ?? '_ignore';
$mapping = $config['column-do-mapping'][$i] ?? false;
// if the column can be mapped depends on the config:
$canMap = (bool)config(sprintf('csv.import_roles.%s.mappable', $role));
$mapping = $mapping && $canMap;
$config['column-do-mapping'][$i] = $mapping;
}
return $config;
}
/**
* Returns false when it's not necessary to map values. This saves time and is user friendly
* (will skip to the next screen).
*
* @param array $config
*
* @return bool
*/
private function isMappingNecessary(array $config): bool
{
$count = $config['column-count'];
$toBeMapped = 0;
for ($i = 0; $i < $count; ++$i) {
$mapping = $config['column-do-mapping'][$i] ?? false;
if (true === $mapping) {
++$toBeMapped;
}
}
return !(0 === $toBeMapped);
}
/**
* Make sure that the examples do not contain double data values.
*/
private function makeExamplesUnique(): void
{
foreach ($this->examples as $index => $values) {
$this->examples[$index] = array_unique($values);
}
}
/**
* if the user has configured specific fixes to be applied, they must be applied to the example data as well.
*
* @param array $row
*
* @return array
*/
private function processSpecifics(array $row): array
{
$config = $this->importJob->configuration;
$specifics = $config['specifics'] ?? [];
$names = array_keys($specifics);
foreach ($names as $name) {
/** @var SpecificInterface $specific */
$specific = app('FireflyIII\Import\Specifics\\' . $name);
$row = $specific->run($row);
}
return $row;
}
/**
* Save the column count in the job. It's used in a later stage.
*
* @return void
*/
private function saveColumCount(): void
{
$config = $this->importJob->configuration;
$config['column-count'] = $this->totalColumns;
$this->repository->setConfiguration($this->importJob, $config);
}
}

View File

@@ -83,10 +83,8 @@ class ConfigureUploadHandler implements ConfigurationInterface
/**
* @param ImportJob $job
*
* @return ConfigurationInterface
*/
public function setJob(ImportJob $job)
public function setJob(ImportJob $job): void
{
$this->importJob = $job;
$this->repository = app(ImportJobRepositoryInterface::class);