<form id="package-edit" class="package_create_form ckan" method="post" 
  py:attrs="{'class':'has-errors'} if errors else {}"
  xmlns:i18n="http://genshi.edgewall.org/i18n"
  xmlns:py="http://genshi.edgewall.org/"
  xmlns:xi="http://www.w3.org/2001/XInclude">
 
<div class="error-explanation" py:if="error_summary">
<h2>Errors in form</h2>
<p>The form contains invalid entries:</p>
<ul>
  <li py:for="key, error in error_summary.items()">${"%s: %s" % (key, error)}</li>
</ul>
</div>
 
<fieldset>
  <legend>Example Basic information</legend>
  <dl>
    <dt><label class="field_req" for="title">Title *</label></dt>
    <dd><input id="title" name="title" type="text" value="${data.get('title', '')}"/></dd>
    <dd class="instructions basic">The title of the data set.</dd>
    <dd class="instructions further">The main subject of the data should be clear. For cross-government data requirements, such as spend data, specify the public body the data belongs to or its geographical coverage, in order to distinguish your data from other similar datasets in data.gov.uk. If the data relates to a period of time, include that in the name, although this would not be appropriate for data which is updated over time. It is not a description - save that for the Abstract element. Do not give a trailing full stop.</dd>
    <dd class="hints">e.g. Payments to suppliers with a value over &pound;500 from Harlow Council</dd>
    <dd class="field_error" py:if="errors.get('title', '')">${errors.get('title', '')}</dd>
 
    <dt><label class="field_req" for="name">Identifier *</label></dt>
    <dd><input id="name" maxlength="100" name="name" type="text" value="${data.get('name', '')}" /></dd>
    <dd class="instructions basic">A public unique identifier for the dataset</dd>
    <dd class="instructions further">It should be roughly readable, with dashes separating words.</dd>
    <dd class="hints">Format: Two or more lowercase alphanumeric, dash (-) or underscore (_) characters. e.g. uk-road-traffic-statistics-2008 or local-authority-spend-over-500-harlow</dd>
    <dd class="field_error" py:if="errors.get('name', '')">${errors.get('name', '')}</dd>
 
    <dt><label class="field_req" for="notes">Abstract *</label></dt>
    <dd><textarea cols="60" id="notes" name="notes" rows="15">${data.get('notes', '')}</textarea></dd>
    <dd class="instructions basic">The main description of the dataset</dd>
    <dd class="instructions further">It is often displayed with the package title. In particular, it should start with a short sentence that describes the data set succinctly, because the first few words alone may be used in some views of the data sets. Here is the place to state if there are any limitations or deficiencies to the data in order to enable users to evaluate the information; even incomplete data may be adequate for some users.</dd>
    <dd class="field_error" py:if="errors.get('notes', '')">${errors.get('notes', '')}</dd>
  </dl>
</fieldset>
 
<fieldset>
  <legend>Example Details</legend>
  <dl>
    <dt><label class="field_opt" for="date_released">Date released</label></dt>
    <dd><input id="date_released" name="date_released" size="40" type="text" value="${data.get('date_released', '')}" /></dd>
    <dd class="instructions basic">The date of the official release of the initial version of the dataset</dd>
    <dd class="instructions further">This is probably not the date that it is uploaded to data.gov.uk. Be careful not to confuse a new 'version' of some data with a new dataset covering another time period or geographic area.</dd>
    <dd class="hints">DD/MM/YYYY</dd>
    <dd class="field_error" py:if="errors.get('date_released', '')">${errors.get('date_released', '')}</dd>
 
    <dt><label class="field_opt" for="date_updated">Date updated</label></dt>
    <dd><input id="date_updated" name="date_updated" size="40" type="text" value="${data.get('date_updated', '')}"/></dd>
    <dd class="instructions basic">The date of release of the most recent version of the dataset</dd>
    <dd class="instructions further">This is not necessarily the date when it was updated on data.gov.uk. As with 'Date released', this is for updates to a particular dataset, such as corrections or refinements, not for that of a new time period.</dd>
    <dd class="hints">DD/MM/YYYY</dd>
    <dd class="field_error" py:if="errors.get('date_updated', '')">${errors.get('date_updated', '')}</dd>
 
    <dt><label class="field_opt" for="date_update_future">Date to be published</label></dt>
    <dd><input id="date_update_future" name="date_update_future" size="40" type="text" value="${data.get('date_update_future', '')}"/></dd>
    <dd class="instructions basic">When the dataset will be updated in the future, if appropriate</dd>
    <dd class="hints">DD/MM/YYYY</dd>
    <dd class="field_error" py:if="errors.get('date_updated', '')">${errors.get('date_updated', '')}</dd>
 
    <dt><label class="field_opt" for="update_frequency">Update frequency</label></dt>
    <dd>
      <select class="short" id="update_frequency" name="update_frequency">
        <py:for each="freq_name, freq_desc in c.update_frequency">
          <option value="${freq_name}" py:attrs="{'selected': 'selected' if data.get('update_frequency', '') == freq_name else None}" >
            ${freq_desc}
          </option>
        </py:for>
      </select>
      <label class="inline" for="update_frequency-other">Other: 
        <input class="medium-width" id="update_frequency-other" name="update_frequency-other" type="text" value="${data.get('update_frequency-other', '')}"/>
      </label>
    </dd>
    <dd class="instructions basic">How frequently the dataset is updated with new versions</dd>
    <dd class="instructions further">For one-off data, use 'never'. For those once updated but now discontinued, use 'discontinued'.</dd>
    <dd class="field_error" py:if="errors.get('update_frequency', '')">${errors.get('update_frequency', '')}</dd>
 
    <dt><label class="field_opt" for="precision">Precision</label></dt>
    <dd><input id="precision" name="precision" size="40" type="text" value="${data.get('precision', '')}"/></dd>
    <dd class="instructions basic">Indicate the level of precision in the data, to avoid over-interpretation.</dd>
    <dd class="hints">e.g. 'per cent to two decimal places' or 'as supplied by respondents'</dd>
 
    <dt><label class="field_opt" for="geographic_granularity">Geographic granularity</label></dt>
    <dd>
      <select class="short" id="geographic_granularity" name="geographic_granularity">
        <py:for each="geog_name, geog_desc in c.geographic_granularity">
          <option value="${geog_name}" py:attrs="{'selected': 'selected' if data.get('geographic_granularity', '') == geog_name else None}" >
            ${geog_desc}
          </option>
        </py:for>
      </select>
      <label class="inline" for="geographic_granularity-other">Other: 
        <input class="medium-width" id="geographic_granularity-other" 
        name="geographic_granularity-other" type="text" value="${data.get('geographic_granularity-other', '')}"/>
      </label>
    </dd>
    <dd class="instructions basic">The lowest level of geographic detail</dd>
    <dd class="instructions further">This should give the lowest level of geographic detail given in the dataset if it is aggregated. If the data is not aggregated, and so the dataset goes down to the level of the entities being reported on (such as school, hospital, or police station), use 'point'. If none of the choices is appropriate or the granularity varies, please specify in the 'other' element.</dd>
    <dd class="field_error" py:if="errors.get('geographic_granularity', '')">${errors.get('geographic_granularity', '')}</dd>
 
    <dt><label class="field_opt" for="geographic_coverage">Geographic coverage</label></dt>
    <dd py:with="geographic_coverage=data.get('geographic_coverage', [])">
      <label for="england">
        <input id="england" name="geographic_coverage" size="40" type="checkbox" 
        value="england" py:attrs="{'checked': 'checked' if 'england' in geographic_coverage else None}"/> England</label>
      <label for="scotland">
        <input id="scotland" name="geographic_coverage" size="40" type="checkbox" 
        value="scotland" py:attrs="{'checked': 'checked' if 'scotland' in geographic_coverage else None}"/> Scotland</label>
      <label for="wales">
        <input id="wales" name="geographic_coverage" size="40" type="checkbox" 
        value="wales" py:attrs="{'checked': 'checked' if 'wales' in geographic_coverage else None}"/> Wales</label>
      <label for="northern_ireland">
        <input id="northern_ireland" name="geographic_coverage" size="40" type="checkbox" 
        value="northern_ireland" py:attrs="{'checked': 'checked' if 'northern_ireland' in geographic_coverage else None}"/> Northern Ireland</label>
      <label for="overseas">
        <input id="overseas" name="geographic_coverage" size="40" type="checkbox" 
        value="overseas" py:attrs="{'checked': 'checked' if 'overseas' in geographic_coverage else None}"/> Overseas</label>
      <label for="global">
        <input id="global" name="geographic_coverage" size="40" type="checkbox" 
        value="global" py:attrs="{'checked': 'checked' if 'global' in geographic_coverage else None}"/> Global</label>
    </dd>
    <dd class="instructions basic">The geographic coverage of this dataset.</dd>
    <dd class="instructions further">Where a dataset covers multiple areas, the system will automatically group these (e.g. 'England', 'Scotland' and 'Wales' all being selected would be shown as 'Great Britain').</dd>
 
    <dt><label class="field_opt" for="temporal_granularity">Temporal granularity</label></dt>
    <dd>
      <select class="short" id="temporal_granularity" name="temporal_granularity">
        <py:for each="temporal_name, temporal_desc in c.temporal_granularity">
          <option value="${temporal_name}" py:attrs="{'selected': 'selected' if data.get('temporal_granularity', '') == temporal_name else None}" >
            ${temporal_name}
          </option>
        </py:for>
      </select>
      <label class="inline" for="temporal_granularity-other">Other: 
        <input class="medium-width" id="temporal_granularity-other" name="temporal_granularity-other" type="text" value="${data.get('temporal_granularity-other', '')}"/>
      </label>
    </dd>
    <dd class="instructions basic">The lowest level of temporal detail granularity</dd>
    <dd class="instructions further">This should give the lowest level of temporal detail given in the dataset if it is aggregated, expressed as an interval of time. If the data is not aggregated over time, and so the dataset goes down to the instants that reported events occurred (such as the timings of high and low tides), use 'point'. If none of the choices is appropriate or the granularity varies, please specify in the 'other' element.</dd>
 
    <dt><label class="field_opt" for="temporal_coverage">Temporal coverage</label></dt>
    <dd>
      <input class="medium-width" id="temporal_coverage-from" name="temporal_coverage-from" type="text" value="${data.get('temporal_coverage-from', '')}" /> - 
      <input class="medium-width" id="temporal_coverage-to" name="temporal_coverage-to" type="text" value