A module of the pyhdf package implementing the VS (Vdata table) API of the NCSA HDF4 library. (see: hdf.ncsa.uiuc.edu)
Version: 0.7-3 Date: July 13 2005
Introduction VS module key features Accessing the VS module Package components Prerequisites Documentation Summary of differences between the pyhdf and C VS API Error handling VS needs support from the HDF module Classes summary Data types Attribute access: low and high level Predefined attributes Record access: low and high level Programming models
Creating and initializing a new vdata Appending records to a vdata Updating records in a vdata Reading a vdataModule documentation
VS is one of the modules composing pyhdf, a python package implementing the NCSA HDF library and letting one manage HDF files from within a python program. Two versions of the HDF library currently exist, version 4 and version 5. pyhdf only implements version 4 of the library. Many different APIs are to be found inside the HDF4 specification. Currently, pyhdf implements just a few of those: the SD, VS and V APIs. Other APIs should be added in the future (GR, AN, etc).
VS allows the definition of structured data tables inside an HDF file. Those tables are designated as “vdatas” (the name has to do with data associated with the “vertices” of geometrical models, the storage of which the API was originally designed for). A vdata is composed of a fixed number of columns (also called fields), where a column can store a fixed number of data values, all of the same type. The number of values allowed inside a field is called the “order” of the field. A table is composed of a varying number of rows (also called records), a record representing the sequence of values stored in each field of the vdata.
A vdata is associated with a descriptive name, and likewise each field of the vdata. A vdata can also be tagged with a “class” to further describe the vdata purpose. Records and fields are identified by a zero-based index. An arbitrary number of attributes of different types can be attached to a vdata as a whole, or to its individual fields. An attribute is a (name, value) pair, where “value” can be of many types, and be either single or multi-valued. The number of values stored in an attribute is called the “order” of the attribute.
The following example illustrates a simple vdata that could be stored inside an HDF file. See section “Programming models” for an example program implementing this vdata.
INVENTORY (experimental status)——————————————i——- partid description qty wght(lb) price($) ————————————————– Q1234 bolt 12 0.01 0.05 B5432 brush 10 0.4 4.25 S7613 scissor 2 0.2 3.75
The vdata is composed of 5 fields. 3 records are shown (of course, a vdata can store much more than that). “INVENTORY” would be the vdata name, and “partid”, “description”, etc, would be the field names. The data type varies between fields. “partid” and “description” would be of “multicharacter” type (aka “string”), “qty” would be a integer, and “wght” and “price” would be floats. The text in parentheses could be stored as attributes. A “status” attribute could be defined for the table as a whole, and given the value “experimental”. Likewise, a “unit” attribute could be associated with fields “wght” and “price”, and given the values “lb” and “$”, resp.
The VS API allows one to create, locate and open a vdata inside an HDF file, update and append records inside it, read records randomly or sequentially, and access and update the vdata and field attributes. Attributes can be read and written using the familiar python “dot notation”, and records can be read and written by indexing and slicing the vdata as if it were a python sequence.
VS key features are as follows.
- -pyhdf implements almost every routine of the original VS API.
- Only a few have been ignored, most of them being of a rare use:
- VSgetblocksize() / VSsetblocksize()
- VSsetnumblocks()
- VSlone
- -It is quite straightforward to go from a C version to a python version
- of a program accessing the VS API, and to learn VS usage by refering to the C API documentation.
- -A few high-level python methods have been developped to ease
- programmers task. Of greatest interest are the following:
- Access to attributes through the familiar “dot notation”.
- Indexing and slicing a vdata to read and write its records, similarly to a python sequence.
- Easy retrieval of info on a vdata and its fields.
- Easy creation of vdatas.
To access the VS module a python program can say one of:
>>> import pyhdf.VS # must prefix names with "pyhdf.VS."
>>> from pyhdf import VS # must prefix names with "VS."
>>> from pyhdf.VS import * # names need no prefix
This document assumes the last import style is used.
VS is not self-contained, and needs functionnality provided by another pyhdf module, namely the HDF module. This module must thus be imported also:
>>> from HDF import *
pyhdf is a proper Python package, eg a collection of modules stored under a directory whose name is that of the package and which stores an __init__.py file. Following the normal installation procedure, this directory will be <python-lib>/site-packages/pyhdf’, where <python-lib> stands for the python installation directory.
For each HDF API exists a corresponding set of modules.
The following modules are related to the VS API.
- _hdfext C extension module responsible for wrapping the HDF
- C library for all python modules
- hdfext python module implementing some utility functions
- complementing the _hdfext extension module
error defines the HDF4Error exception HDF python module providing support to the VS module VS python module wrapping the VS API routines inside
an OOP framework
_hdfext and hdfext were generated using the SWIG preprocessor. SWIG is however not needed to run the package. Those two modules are meant to do their work in the background, and should never be called directly. Only HDF and VS should be imported by the user program.
The following software must be installed in order for VS to work.
- HDF (v4) library
pyhdf does not include the HDF4 library, which must be installed separately.
HDF is available at: “http://hdf.ncsa.uiuc.edu/obtain.html”.
Numeric is also needed by the SD module. See the SD module documentation.
pyhdf has been written so as to stick as closely as possible to the naming conventions and calling sequences documented inside the “HDF User s Guide” manual. Even if pyhdf gives an OOP twist to the C API, the manual can be easily used as a documentary source for pyhdf, once the class to which a function belongs has been identified, and of course once requirements imposed by the Python langage have been taken into account. Consequently, this documentation will not attempt to provide an exhaustive coverage of the HDF VS API. For this, the user is referred to the above manual. The documentation of each pyhdf method will indicate the name of the equivalent routine as it is found inside the C API.
This document (in both its text and html versions) has been completely produced using “pydoc”, the Python documentation generator (which made its debut in the 2.1 Python release). pydoc can also be used as an on-line help tool. For example, to know everything about the VS.VD class, say:
>>> from pydoc import help
>>> from pyhdf.VS import *
>>> help(VD)
To be more specific and get help only for the read() method of the VD class:
>>> help(VD.read)
pydoc can also be called from the command line, as in:
% pydoc pyhdf.VS.VD # doc for the whole VD class % pydoc pyhdf.VS.VD.read # doc for the VD.read method
Most of the differences between the pyhdf and C VS API can be summarized as follows.
- -In the C API, every function returns an integer status code, and values
- computed by the function are returned through one or more pointers passed as arguments.
- -In pyhdf, error statuses are returned through the Python exception
- mechanism, and values are returned as the method result. When the C API specifies that multiple values are returned, pyhdf returns a sequence of values, which are ordered similarly to the pointers in the C function argument list.
All errors reported by the C VS API with a SUCCESS/FAIL error code are reported by pyhdf using the Python exception mechanism. When the C library reports a FAIL status, pyhdf raises an HDF4Error exception (a subclass of Exception) with a descriptive message. Unfortunately, the C library is rarely informative about the cause of the error. pyhdf does its best to try to document the error, but most of the time cannot do more than saying “execution error”.
The VS module is not self-contained (countrary to the SD module). It requires help from the HDF module, namely:
- -the HDF.HDF class to open and close the HDF file, and initialize the
- VS interface
- -the HDF.HC class to provide different sorts of constants (opening modes,
- data types, etc).
A program wanting to access HDF vdatas will almost always need to execute the following minimal set of calls:
>>> from pyhdf.HDF import *
>>> from pyhdf.VS import *
>>> hdfFile = HDF(name, HC.xxx)# open HDF file
>>> vs = hdfFile.vstart() # initialize VS interface on HDF file
>>> ... # manipulate vdatas through "vs"
>>> vs.end() # terminate VS interface
>>> hdfFile.close() # close HDF file
pyhdf wraps the VS API using different python classes:
VS HDF VS interface VD vdata VDField vdata field VDattr attribute (either at the vdata or field level)
In more detail:
- VS The VS class implements the VS (Vdata) interface applied to an
HDF file. This class encapsulates the hdf instance, and all the top-level functions of the VS API.
To create a VS instance, call the vstart() method of an HDF instance.
- methods:
- constructors:
- attach() open an existing vdata given its name or
- reference number, or create a new one, returning a VD instance
- create() create a new vdata and define its structure,
- returning a VD instance
- creating and initializing a simple vdata
- storedata() create a single-field vdata and initialize
- its values
- closing the interface
- end() close the VS interface on the HDF file
- searching
find() get a vdata reference number given its name next() get the reference number of the vdata following
a given one- inquiry
- vdatainfo() return info about all the vdatas in the
- HDF file
- VD The VD class describes a vdata. It encapsulates
the VS instance to which the vdata belongs, and the vdata identifier.
To instantiate a VD class, call the attach() or create() method of a VS class instance.
- methods:
- constructors
- attr() create a VDAttr instance representing a
- vdata attribute; “dot notation” can also be used to access a vdata attribute
- field() return a VDField instance representing a given
- field of the vdata
- closing vdata
- detach() end access to the vdata
- defining fields
fdefine() define the name, type and order of a new field setfields() define the field names and field order for
the read() and write() methods; also used to initialize the structure of a vdata previously created with the VS.attach() method- reading and writing
- read() return the values of a number of records
- starting at the current record position
seek() reset the current record position seekend() seek past the last record tell() return the current record position write() write a number of records starting at the
current record position- inquiry
attrinfo() return info about all the vdata attributes fexist() check if a vdata contains a given set of fields fieldinfo() return info about all the vdata fields findattr() locate an attribute, returning a VDAttr instance
if foundinquire() return info about the vdata sizeof() return the size in bytes of one or more fields
- VDField The VDField class represents a vdata field. It encapsulates
the VD instance to which the field belongs, and the field index number.
To instantiate a VDField, call the field() method of a VD class instance.
- methods:
- constructors:
- attr() return a VDAttr instance representing an
- attribute of the field; “dot notation” can also be used to get/set an attribute.
- inquiry
attrinfo() return info about all the field attributes find() locate an attribute, returning a VDAttr
instance if found- VDAttr The VDAttr class encapsulates methods used to set and query
attributes defined at the level either of the vdata or the vdata field.
To create an instance of this class, call the attr() or findattr() methods of a VD instance (for vdata attributes), or call the attr() or find() methods of a VDField instance (for field attributes).
- methods:
- get / set
- get() get the attribute value set() set the attribute value
- info
- info() retrieve info about the attribute
Data types come into play when first defining vdata fields and attributes, and later when querying the definition of those fields and attributes. Data types are specified using the symbolic constants defined inside the HC class of the HDF module.
- CHAR and CHAR8 (equivalent): an 8-bit character.
- UCHAR, UCHAR8 and UINT8 (equivalent): unsigned 8-bit values (0 to 255)
- INT8: signed 8-bit values (-128 to 127)
- INT16: signed 16-bit values
- UINT16: unsigned 16 bit values
- INT32: signed 32 bit values
- UINT32: unsigned 32 bit values
- FLOAT32: 32 bit floating point values (C floats)
- FLOAT64: 64 bit floating point values (C doubles)
There is no explicit “string” type. To simulate a string, set the field or attribute type to CHAR, and set the field or attribute “order” to a value of ‘n’ > 1. This creates and “array of characters”, close to a string (except that strings will always be of length ‘n’, right-padded with spaces if necessary).
The VS API allow setting attributes on vdatas and vdata fields. Attributes can be of many types (int, float, char) of different bit lengths (8, 16, 32, 64 bits), and can be single or multi-valued. Values of a multi-valued attribute must all be of the same type.
Attributes can be set and queried in two different ways. First, given a VD instance (describing a vdata object) or a VDField instance (describing a vdata field), the attr() method of that instance is called to create a VDAttr instance representing the wanted attribute (possibly non existent). The set() method of this VDAttr instance is then called to define the attribute value, creating it if it does not already exist. The get() method returns the current attribute value. Here is an example.
>>> from pyhdf.HDF import *
>>> from pyhdf.VS import *
>>> f = HDF('test.hdf', HC.WRITE) # Open file 'test.hdf' in write mode
>>> vs = f.vstart() # init vdata interface
>>> vd = vs.attach('vtest', 1) # attach vdata 'vtest' in write mode
>>> attr = vd.attr('version') # prepare to define the 'version' attribute
# on the vdata
>>> attr.set(HC.CHAR8,'1.0') # set attribute 'version' to string '1.0'
>>> print attr.get() # get and print attribute value
>>> fld = vd.field('fld1') # obtain a field instance for field 'fld1'
>>> attr = fld.attr('range') # prepare to define attribute 'range' on
# this field
>>> attr.set(HC.INT32,(-10, 15)) # set attribute 'range' to a pair of ints
>>> print attr.get() # get and print attribute value
>>> vd.detach() # "close" the vdata
>>> vs.end() # terminate the vdata interface
>>> f.close() # close the HDF file
The second way consists of setting/querying an attribute as if it were a normal python class attribute, using the usual dot notation. Above example then becomes:
>>> from pyhdf.HDF import *
>>> from pyhdf.VS import *
>>> f = HDF('test.hdf', HC.WRITE) # Open file 'test.hdf' in write mode
>>> vs = f.vstart() # init vdata interface
>>> vd = vs.attach('vtest', 1) # attach vdata 'vtest' in write mode
>>> vd.version = '1.0' # create vdata attribute 'version',
# setting it to string '1.0'
>>> print vd.version # print attribute value
>>> fld = vd.field('fld1') # obtain a field instance for field 'fld1'
>>> fld.range = (-10, 15) # create field attribute 'range', setting
# it to the pair of ints (-10, 15)
>>> print fld.range # print attribute value
>>> vd.detach() # "close" the vdata
>>> vs.end() # terminate the vdata interface
>>> f.close() # close the HDF file
Note how the dot notation greatly simplifies and clarifies the code. Some latitude is however lost by manipulating attributes in that way, because the pyhdf package, not the programmer, is then responsible of setting the attribute type. The attribute type is chosen to be one of:
HC.CHAR8 if the attribute value is a string HC.INT32 if all attribute values are integers HC.FLOAT64 otherwise
The first way of handling attribute values must be used if one wants to define an attribute of any other type (for ex. 8 or 16 bit integers, signed or unsigned). Also, only a VDAttr instance gives access to attribute info, through its info() method.
However, accessing HDF attributes as if they were python attributes raises an important issue. There must exist a way to assign generic attributes to the python objects without requiring those attributes to be converted to HDF attributes. pyhdf uses the following rule: an attribute whose name starts with an underscore (‘_’) is either a “predefined” attribute (see below) or a standard python attribute. Otherwise, the attribute is handled as an HDF attribute. Also, HDF attributes are not stored inside the object dictionnary: the python dir() function will not list them.
Attribute values can be updated, but it is illegal to try to change the value type, or the attribute order (number of values). This is important for attributes holding string values. An attribute initialized with an ‘n’ character string is simply a character attribute of order ‘n’ (eg a character array of length ‘n’). If ‘vd’ is a vdata and we initialize its ‘a1’ attribute as ‘vd.a1 = “abcdef”’, then a subsequent update attempt like ‘vd.a1 = “12”’ will fail, because we then try to change the order of the attribute (from 6 to 2). It is mandatory to keep the length of string attributes constant. Examples below show simple ways how this can be done.
The VD and VDField classes support predefined attributes to get (and occasionnaly set) attribute values easily, without having to call a class method. The names of predefined attributes all start with an underscore (‘_’).
In the following tables, the RW column holds an X if the attribute is read/write. See the HDF User s guide for details about more “exotic” topics like “class”, “faked vdata” and “tag”.
VD predefined attributes
_class X class name VSgetclass/VSsetclass _fields list of field names VSgetfields _interlace X interlace mode VSgetinterlace/VSsetinterlace _isattr true if vdata is “faked” VSisattr
by HDF to hold attributes_name X name of the vdata VSgetname/VSsetname _nattrs number of attributes VSfnattrs _nfields number of fields VFnfields _nrecs number of records VSelts _recsize record size (bytes) VSQueryvsize _refnum reference number VSQueryref _tag vdata tag VSQuerytag _tnattrs total number of vdata and VSnattrs
field attributesVDField predefined attributes
_esize external size (bytes) VFfieldesize _index index number VSfindex _isize internal size (bytes) VFfieldisize _name name VFfieldname _nattrs number of attributes VSfnattrs _order order (number of values) VFfieldorder _type field type (HC.xxx) VFfieldtype
vdata records can be read and written in two different ways. The first one consists of calling the basic I/O methods of the vdata:
- seek() to set the current record position, if necessary;
- read() to retrieve a given number of records from that position;
- write() to write a given number of records starting at that position
A second, higher level way, lets one see a vdata similarly to a python sequence, and access its contents using the familiar indexing and slicing notation in square brackets. Reading and writing a vdata as if it were a python sequence may often look simpler, and improve code legibility.
Here are some examples of how a vdata ‘vd’ holding 3 fields could be read.
>>> print vd[0] # print record 0
>>> print vd[-1] # print last record
>>> print vd[2:] # print records 2 and those that follow
>>> print vd[:] # print all records
>>> print vd[:,0] # print field 0 of all records
>>> print vd[:3,:2] # print first 2 fields of first 3 records
As the above examples show, the usual python rules are obeyed regarding the interpretation of indexing and slicing values. Note that the vdata fields can be indexed and sliced, not only the records. The setfields() method can also be used to select a subset to the vdata fields (setfields() also let you reorder the fields). When the vdata is indexed (as opposed to being sliced), a single record is returned as a list of values. When the vdata is sliced, a list of records is always returned (thus a 2-level list), even if the slice contains only one record.
A vdata can also be written similarly to a python sequence. When indexing the vdata (as opposed to slicing it), a single record must be assigned, and the record must be given as a sequence of values. It is legal to use as an index the current number of records in the vdata: the record is then appended to the vdata. When slicing the vdata, the records assigned to the slice must always be given as a list of records, even if only one record is assigned. Also, the number of records assigned must always match the width of the slice, except if the slice includes or goes past the last record of the vdata. In that case, the number of records assigned can exceed the width of the slice, and the extra records are appended to the vdata. So, to append records to vdata ‘vd’, simply assign records to the slice ‘vd[vd._nrecs:]’. Note that, even if the ‘field’ dimension can be specified in the left-hand side expression, there is no real interest in doing so, since all fields must be specified when assigning a record to the vdata: it is an error to try to assign just a few of the fields.
For example, given a vdata ‘vd’ holding 5 records, and lists ‘reca’, ‘recb’, etc, holding record values:
vd[0] = reca # updates record 0 vd[0,:] = reca # specifying fields is OK, but useless vd[0,1:] = reca[1:] # error: all fields must be assigned vd[1] = [recb, recc] # error: only one record allowed vd[5] = recc # append one record vd[1:3] = [reca,recb] # updates second and third record vd[1:4] = [reca, recb] # error: 3 records needed vd[5:] = [reca,recb] # appends 2 records to the vdata vd[4:] = [reca, recb] # updates last record, append one
The following code can serve as a model for the creation and initialization of a new vdata. It implements the INVENTORY example described in the “Introduction” section.
from pyhdf.HDF import * from pyhdf.VS import *
# Open HDF file and initialize the VS interface f = HDF(‘inventory.hdf’, # Open file ‘inventory.hdf’ in write mode
HC.WRITE|HC.CREATE) # creating it if it does not existvs = f.vstart() # init vdata interface
# Create vdata and define its structure vd = vs.create( # create a new vdata
- ‘INVENTORY’, # name of the vdata
- # fields of the vdata follow
- ((‘partid’,HC.CHAR8, 5), # 5 char string
- (‘description’,HC.CHAR8, 10), # 10 char string field (‘qty’,HC.INT16, 1), # 1 16 bit int field (‘wght’,HC.FLOAT32, 1), # 1 32 bit float (‘price’,HC.FLOAT32,1) # 1 32 bit float
)) # 5 fields allocated in the vdata
# Set attributes on the vdata and its fields vd.field(‘wght’).unit = ‘lb’ vd.field(‘price’).unit = ‘$’ # In order to be able to update a string attribute, it must # always be set to the same length. This sets ‘status’ to a 20 # char long, left-justified string, padded with spaces on the right. vd.status = “%-20s” % ‘phase 1 done’
# Store records vd.write(( # write 3 records
(‘Q1234’, ‘bolt’,12, 0.01, 0.05), # record 1 (‘B5432’, ‘brush’, 10, 0.4, 4.25), # record 2 (‘S7613’, ‘scissor’, 2, 0.2, 3.75) # record 3 ))vd.detach() # “close” the vdata
vs.end() # terminate the vdata interface f.close() # close the HDF file
Note that is mandatory to always write whole records to the vdata. Note also the comments about the initialization of the ‘status’ vdata attribute. We want to be able update this attribute (see following examples). However, the VS API prohibits changing an attribute type when updating its value. Since the length (order) of an attribute is part of its type, we make sure of setting the attribute to a length long enough to accomodate the longest possible string we migh want to assign to the attribute.
Appending records requires first seeking to the end of the vdata, to avoid overwriting existing records. The following code can serve as a model. The INVENTORY vdata created before is used.
from pyhdf.HDF import * from pyhdf.VS import *
- f = HDF(‘inventory.hdf’, # Open ‘inventory.hdf’ in write mode
- HC.WRITE|HC.CREATE) # creating it if it does not exist
vs = f.vstart() # init vdata interface vd = vs.attach(‘INVENTORY’, 1) # attach ‘INVENTORY’ in write mode
# Update the `status’ vdata attribute. The attribute length must not # change. We call the attribute info() method, which returns a list # where number of values (eg string length) is stored at index 2. # We then assign a left justified string of exactly that length. len = vd.attr(‘status’).info()[2] vd.status = ‘%-*s’ % (len, ‘phase 2 done’)
- vd[vd._nrecs:] = ( # append 2 records
(‘A4321’, ‘axe’, 5, 1.5, 25), # first record (‘C3214’, ‘cup’, 100, 0.1, 3.25) # second record
)vd.detach() # “close” the vdata
vs.end() # terminate the vdata interface f.close() # close the HDF file
Note how, when updating the value of the ‘status’ vdata attribute, we take care of assigning a value of the same length as that of the original value. Otherwise, the assignment would raise an exception. Records are written by assigning the vdata through a slicing expression, like a python sequence. By specifying the number of records as the start of the slice, the records are appended to the vdata.
Updating requires seeking to the record to update before writing the new records. New data will overwrite this record and all records that follow, until a new seek is performed or the vdata is closed. Note that record numbering starts at 0.
The following code can serve as a model. The INVENTORY vdata created before is used.
from pyhdf.HDF import * from pyhdf.VS import *
- f = HDF(‘inventory.hdf’, # Open ‘inventory.hdf’ in write mode
- HC.WRITE|HC.CREATE) # creating it if it does not exist
vs = f.vstart() # init vdata interface vd = vs.attach(‘INVENTORY’, 1) # attach ‘INVENTORY’ in write mode
# Update the `status’ vdata attribute. The attribute length must not # change. We call the attribute info() method, which returns a list # where number of values (eg string length) is stored at index 2. # We then assign a left justified string of exactly that length. len = vd.attr(‘status’).info()[2] vd.status = ‘%-*s’ % (len, ‘phase 3 done’)
# Update record at index 1 (second record) vd[1] = (‘Z4367’, ‘surprise’, 10, 3.1, 44.5) # Update record at index 4, and all those that follow vd[4:] = (
(‘QR231’, ‘toy’, 12, 2.5, 45), (‘R3389’, ‘robot’, 3, 45, 2000) )vd.detach() # “close” the vdata vs.end() # terminate the vdata interface f.close() # close the HDF file
The following example shows how read the vdata attributes and sequentially maneuver through its records. Note how we use the exception mechanism to break out of the reading loop when we reach the end of the vdata.
from pyhdf.HDF import * from pyhdf.VS import *
f = HDF(‘inventory.hdf’) # open ‘inventory.hdf’ in read mode vs = f.vstart() # init vdata interface vd = vs.attach(‘INVENTORY’) # attach ‘INVENTORY’ in read mode
# Display some vdata attributes print “status:”, vd.status print “vdata: ”, vd._name # predefined attribute: vdata name print “nrecs: ”, vd._nrecs # predefined attribute: num records
# Display value of attribute ‘unit’ for all fields on which # this attribute is set print “units: ”, for fieldName in vd._fields: # loop over all field names
- try:
- # instantiate field and obtain value of attribute ‘unit’ v = vd.field(fieldName).unit print “%s: %s” % (fieldName, v),
- except: # no ‘unit’ attribute: ignore
- pass
print “” print “”
# Display table header. header = “%-7s %-12s %3s %4s %8s” % tuple(vd._fields) print “-” * len(header) print header print “-” * len(header)
# Loop over the vdata records, displaying each record as a table row. # Current record position is 0 after attaching the vdata. while 1:
- try:
- # rec = vd[vd.tell()]
- print “%-7s %-12s %3d %4.1f %8.2f” % tuple(rec[0])
- except HDF4Error: # end of vdata reached
- break
vd.detach() # “close” the vdata vs.end() # terminate the vdata interface f.close() # close the HDF file
In the previous example, the reading/displaying loop can be greatly simplified by rewriting it as follows:
from pyhdf.HDF import * from pyhdf.VS import *
f = HDF(‘inventory.hdf’) # open ‘inventory.hdf’ in read mode vs = f.vstart() # init vdata interface vd = vs.attach(‘INVENTORY’) # attach ‘INVENTORY’ in read mode
# Read all records at once, and loop over the sequence. for rec in vd[:]:
print “%-7s %-12s %3d %4.1f %8.2f” % tuple(rec)vd.detach() # “close” the vdata ...
The indexing expression ‘vd[:]’ returns the complete set of records, which can then be looped over using a ‘for’ statement. This style of loop is quite clean, and should look very familiar to python adepts.
The VS class implements the VS (Vdata) interface applied to an HDF file. To instantiate a VS class, call the vstart() method of an HDF instance.
Locate an existing vdata or create a new vdata in the HDF file, returning a VD instance.
C library equivalent : VSattach
After creating a new vdata (num_name == -1), fields must be defined using method fdefine() of the VD instance, and those fields must be allocated to the vdata with method setfields(). Same results can be achieved, but more simply, by calling the create() method of the VS instance.
Create a new vdata, setting its name and allocating its fields.
name Name to assign to the vdata fields Sequence of field definitions. Each field definition
- is a sequence with the following elements in order:
- -field name -field type (one of HC.xxx constants) -field order (number of values)
Fields are allocated to the vdata in the given order
write mode
vd._name = name, to set the vdata name
each field
vd.setfields(...), to allocate fields to the vdata
C library equivalent : no equivalent
Close the VS interface.
C library equivalent : Vend
Get the reference number of a vdata given its name. The vdata can then be opened (attached) by passing this reference number to the attach() method.
C library equivalent : VSfind
Get the reference number of the vdata following a given vdata.
An exception is raised if no vdata follows the one given by vRef.
C library equivalent : VSgetid
Create and initialize a single field vdata, returning the vdata reference number.
fieldName Name of the single field in the vadata to create values Sequence of values to store in the field;. Each value can
itself be a sequence, in which case the field will be multivalued (all second-level sequences must be of the same length)
vName Name of the vdata to create vClass Vdata class (string)
C library equivalent : VHstoredata / VHstoredatam
Return info about all the file vdatas.
List of vdata descriptions. Each vdata is described as a 9-element tuple, composed of the following:
-vdata name -vdata class -vdata reference number -vdata number of records -vdata number of fields -vdata number of attributes -vdata record size in bytes -vdata tag number -vdata interlace mode
C library equivalent : no equivalent
Close the VS interface.
C library equivalent : Vend
The VD class encapsulates the functionnality of a vdata. To instantiate a VD class, call the attach() or the create() method of a VS class instance.
Create a VDAttr instance representing a vdata attribute.
C library equivalent : no equivalent
Return info about all the vdata attributes.
dictionnary describing each vdata attribute; for each attribute a (name,data) pair is added to the dictionary, where ‘data’ is a tuple holding:
-attribute data type (one of HC.xxx constants) -attribute order -attribute value -attribute size in bytes
C library equivalent : no equivalent
Terminate access to the vdata.
C library equivalent : VSdetach
Define a field. To initialize a newly created vdata with fields created with fdefine(), assign a tuple of field names to the _fields attribute or call the setfields() method.
C library equivalent : VSfdefine
Check if a vdata contains a given set of fields.
C library equivalent : VSfexist
Get a VDField instance representing a field of the vdata.
C library equivalent : no equivalent
Retrieve info about all vdata fields.
list where each element describes a field of the vdata; each field is described by an 7-element tuple containing the following elements:
-field name -field data type (one of HC.xxx constants) -field order -number of attributes attached to the field -field index number -field external size -field internal size
C library equivalent : no equivalent
Search the vdata for a given attribute.
if found, VDAttr instance describing the attribute None otherwise
C library equivalent : VSfindattr
Retrieve info about the vdata.
C library equivalent : VSinquire
Retrieve the values of a number of records, starting at the current record position. The current record position is advanced by the number of records read. Current position is 0 after “opening” the vdata with the attach() method.
An exception is raised if the current record position is already at the end of the vdata when read() is called. This exception can be caught as an “end of vdata” indication to exit a loop which scans each record of the vdata. Otherwise, the number of records to be read is lowered to the number of records remaining in the vdata, if that number is less than the number asked for by parameter ‘nRec’. Setting ‘nRec’ to an arbitrarily large value can thus be used to retrieve the remaining records in the vdata.
C library equivalent : VSread
Seek to the beginning of the record identified by its record index. A succeeding read will load this record in memory.
An exception is raised if an attempt is made to seek beyond the last record.
The C API prohibits seeking past the next-to-last record, forcing one to read the last record to advance to the end of the vdata. The python API removes this limitation.
Seeking to the end of the vdata can also be done by calling method `seekend()’.
C library equivalent : VSseek
Set the current record position past the last vdata record. Subsequent write() calls will append records to the vdata.
C library equivalent : no equivalent
Define the name and order of the fields to access with the read() and write() methods.
C library equivalent : VSsetfields
setfields() indicates how to perform the matching between the vdata fields and the values passed to the write() method or returned by the read() method.
For example, if the vdata contains fields ‘a’, ‘b’ and ‘c’ and a “setfields(‘c’,’a’)” call is made, read() will thereafter return for each record the values of field ‘c’ and ‘a’, in that order. Field ‘b’ will be ignored.
When writing to a vdata, setfields() has a second usage. It is used to initialize the structure of the vdata, that is, the name and order of the fields that it will contain. The fields must have been previously defined by calls to the fdefine() method. Following that first call, setfields() can be called again to change the order in which the record values will be passed to the write() method. However, since it is mandatory to write whole records, subsequent calls to setfields() must specify every field name: only the field order can be changed.
Retrieve the size in bytes of the given fields.
C library equivalent : VSsizeof
Return current record position in the vdata.
C library equivalent : no equivalent
Write records to the vdata. Writing starts at the current record position, which is advanced by the number of records written.
To append to a vdata already holding ‘n’ records, it is necessary to first move the current record position to ‘n-1’ with a call to method seek(), then to call method read() for the side effect of advancing the current record position past this last record. Method seekend() does just that.
C library equivalent : VSwrite
The VDField class represents a vdata field. To create a VDField instance, call the field() method of a VD class instance.
Create a VDAttr instance representing a field attribute.
C library equivalent : no equivalent
Return info about all the field attributes.
dictionnary describing each vdata attribute; for each attribute a (name,data) pair is added to the dictionary, where ‘data’ is a tuple holding:
-attribute data type (one of HC.xxx constants) -attribute order -attribute value -attribute size in bytes
C library equivalent : no equivalent
Search the field for a given attribute.
if found, VDAttr instance describing the attribute None otherwise
C library equivalent : VSfindattr
The VDAttr class encapsulates methods used to set and query attributes defined at the level either of the vdata or of the vdata field. To create an instance of this class, call the attr() method of a VD (vdata) or VDField (vdata field) instance.
Retrieve the attribute value.
C library equivalent : VSgetattr
Retrieve info about the attribute.
C library equivalent : VSattrinfo
Set the attribute value.
data_type : attribute data type (see constants HC.xxx) values : attribute value(s); specify a list to create
a multi-valued attribute; a string valued attribute can be created by setting ‘data_type’ to HC.CHAR8 and ‘values’ to the corresponding string
If the attribute already exists, it will be updated. However, it is illegal to try to change its data type or its order (number of values).
C library equivalent : VSsetattr