kitwallace / treemap Goto Github PK

Add a new layer for sites

These sometimes coincide with collections
Gather data from
https://www.bristol.gov.uk/museums-parks-sports-culture/parks-and-estates
http://www.woodlandtrust.org.uk/visiting-woods/
whereelse?

allow update esp via the mobile version

Bought bristoltrees.space cheap for £0.90 from 123-reg - full price is £9
apache2 configuration setup with new address.

needs review and to be generated from select trees

More spacious, large layout

Use fonts to match the BTF website
Headings Merriweather Sans
Body Gentium Book Basic

These are Google fonts - not sure how to use these with existdb

Data collected by Dave Cemlyn

parsing is working ok but there are a number of mistakes in the parse which need to be fixed by editing the base text

Like the one for gloucester road, where the nearest location can be identified

It would be useful to have a list of people as contacts in organisations

Only problem is how to control access to this function - separate script with access controlled via basic authentication

In the data, this is typically shown in the indexable form,

eg Oak, Turkey
rather than Turkey Oak

I've used the supplied name in lists but in expanded entries Ive reversed the name to its normal form

There are some common names which are inconsistantly in the normal form already and should be changed

The reversed parts to get the normal form fails in a few cases eg
Ash, Manna or Flowering Ash
I could split the name on or to add an alternate name

variety/cultivar according to https://en.wikipedia.org/wiki/Wikipedia:Naming_conventions_(flora)
variety should be preceeded by 'var' eg
Populus nigra var italica
cultivar single quoted e.g.
Fraxinus excelsior 'Pendula'

but
Fagus sylvatica pendula
is a cultivar
in the data these forms appear but most are of the form eg
Cornus controversa variegata
which seems to be a cultivar

Collecting the data for the square is useful in its own right since we, the Morley Square Residents are responsible for the upkeep of the trees. Currently we lack an accurate inventory of the trees, their location, species, age and origin.

I will also trial the use of Google earth to collect locations and formated descriptions to capture the data

Odd Unicode characters in the base document eg U2019 are not converted properly by util:binary-to-string and need to be replaced

parser needs to ignore the preamble

Aesculs flava to Aesculus flava
Aesculus turbinate to Aesculus turbinata
Betula ermani to Betula ermanii
73,74 Black Moris negra - missing comma and sb Morus nigra
Catalpa bignoniodes to Catalpa bignonioides
Cedrus deodora sb Cedrus deodara
Cercidifolium japonicum to cercidiphyllum japonicum
Lawson's Cypress to Lawson Cypress
Chinese Ehretia = probably Ehretia acuminata common name Koda
86 missing comma
56 , 57 missing comma
193 missing comma
107 Crypotomeria japonica to Cryptomeria japonica
96 Paulonia tomentosa to Paulownia tomentosa
3 Cupressus Macrocarpa to Cupressus macrocarpa
98 Eucalyptus gunii to Eucalyptus gunnii

RB-35 - correct to 51.460344°, -2.626544°
RB-134 = correct to 51.459069°, -2.626517°

need to position map below the text - but how to know its a mobile phone?

mark to supply

Navigation between tree locate and the tree editor needs to be improved, perhaps by combining the scripts but this will require login rather than basic authentication

Species are not first class entities in this system so there is nowhere to add species data such as Family name. A species dataset could be constructed however using the Latin name

Would have to consider how varieties, cultivars and hybrids where represented

It would be nice to find all trees within a given range of a target tree, ranked by distance

Selection is currently by exact match but it would be better to support hierarchical search so that we can search for either:
trees with only genus defined ie. node
trees in that genus with species defined ie. node and all children
similarly for a specific cultivar or all cultivars in a species
basic indexing serves the node search and free text indexing for the children

Data on TPOs is supposedly available from data.gov.uk form the BCC inspire GIS system but not at all obvious how to do this.

BCC GIS Support team and Katherine Rooney at opendata.bristol contacted to request access -
awaiting further response

to:
Veteran
Champion
Remarkable

http://maps.bristol.gov.uk/pinpoint/ filtered on Environment and Planning - Trees - Bristol University only shows tree at certain locations, but a * search lists 'the first 1500 matching records...' elsewhere (e.g. Royal Fort Garden), even though they are not shown on the map layer.

The dataset cannot be copied of the site. maybe the data is on a separate dataset in https://opendata.bristol.gov.uk/Environment/?

Currently using a hack involving replace and tokenize to extract the text matching a pattern.

https://www.w3.org/TR/xpath-functions-30/#func-analyze-string

Add space between names in BrandonHill & AshtonCourt Collections

Need to be able to get back to the list of selected trees - eg when list is a search. back button works of course but needs to be explicit I think

need indexes:
Species - index latin and common names
Trees - index id , free-text indexes on name, latin, text

mod code to use the free text indexes

Get hold of the data for bristol from the Woodlands Trust ancient Tree Hunt

http://www.ancient-tree-hunt.org.uk/

There are 196 trees recorded in Bristol

Open the Stumps Collection
sort by Id
Select either of the top two stumps listed - with no OBJECTID No. listed
The Search page is displayed - the stump page cannot be opened

Automate the review of the data quality covering
latin names on Species list
common names on species list
geo-coded (lat-long)
..

also add redirections for plane-distance

Need to be able to edit tree data in the database. Note that this will render the database version the master version. Document can be re-exported as xml or csv but not necessarily in the same format as the import format

need to be able to search a collection - should lists be able to be restricted by tag? add tag to search form

Need to replace the query string by a URL language
search to use Post

needs some changes to Apache, the url decoded module and dispatcher

export selected trees as xml, kml or csv - will help with audits

Need a new icon for the desktop and favicon - use of the BTF icon is confusing

Allow user to filter results by SITE_NAME a filtered drop-down or freetext search
Report also Crown Height & Crown Width

The location text would be useful but is difficult to parse since it is often missing and its difficult to distinguish a location sentence from a description.

Also the locations in the text contain sub locations which would need to be lost in order to produce a useful field for selection.

Not sure how these should be handled

We have two descriptions of the trees in St Andrews Park. It would be useful to be able to reconcile this data but its a hard task to find the best match based on proximity between two trees.

Compass direction is still wrong

Not easy to include in the parser but the number ranges could be used
Section A 1-51 veteran trees
Section B 52-182 champion trees
Section C 183-202 remarkable trees

Checking the range is made trickier by the a/b suffixes - suggest change to .1, .2 instead
Add tag to the entry and provide selection by tag