Author Archives: Alejo

find_unum: How to Quickly Find a U-number in the Database

Posted on by
1

The unique identifier of a target in the BDNYC database is its U-number. U-numbers are (almost) completely randomly assigned, and so there is no way of knowing a target’s U-number a priori. Kelle’s Access database is the only place where you can get a target’s U-number… until now.

The method find_unum in the BDNYC database is a quick way to find the U-number of a target. The method, of course, can be used only for targets that already exist in the database.

How you can use find_unum

You can find a U-number by right ascension (ra), declination, and/or any known name. The method gives you the flexibility to enter only one of these if you prefer, and it will print in your terminal the attributes of all targets matching your input.

Let’s look at an example. You need to find the U-number of a target with ra = 10 11 00.2. This is how you do it:

> bdnyc.find_unum(ra=’10 11 00.2′)

The output will be:

—–Potential match 1: —–
[‘unum’, ‘U10885’]
[‘index’, 286]
[‘name’, ‘TVLM 263-71765’]
[‘sptype’, ‘M7.5’]
[‘ra’, ’10 11 00.2′]
[‘dec’, ‘+42 45 03’]
[‘standard’, ‘No’]

Alternatively, instead of ’10 11 00.2′, you could enter ’10 11′, and the method will you give back all targets that have ra starting with ’10 11′.

If you prefer to get the method’s output not as a terminal print-out but as an object, then set the parameter dump as True. The output will be a Python list.

This is the complete method:

bdnyc.find_unum(ra=None, dec=None, name=None, dump=False)

 

 

BDNYC Database Documentation

Posted on by
Reply

Now that the database infrastructure is more complete, I have generated the Sphinx documentation for all the database code. You can find it at

bdnyc.org/database

To have a full understanding of the database, you need both the Sphinx documentation and the Python Database Structure diagram, which is shared in Google Drive.

show_data: Quickly See What Data Exists in the BDNYC Database

Posted on by
3

In the BDNYC database code, I added a new method called show_data. Its purpose is to show you all and any kind of data that exists in the database for a specific target. The method is part of the BDNYC.py module, under the BDNYCData class.

The only input it needs is the U-number of the target. So, for example, if you want to know what data exists for target U10276, type:

> bdnyc.show_data(‘U10276’)

assuming that bdnyc is where you loaded the database using the pickle.load(f) command. The method will automatically check whether the target exists in the database, and if it does, it will print out a list of rows, each one describing a piece of data for the target in question. For the case of U10276, you will get the following in your terminal:

[‘unum’, ‘U10276’]
[‘index’, 74]
[‘name’, ‘sd0423’]
[‘sptype’, ‘T0’]
[‘ra’, ’04 23 48.6′]
[‘dec’, ‘-04 14 04.0’]
[‘standard’, ‘Yes’]
[0, ‘opt’, ‘med’, ‘CT 4m’, ‘2002jan25’]
[1, ‘nir’, ‘high’, ‘NIRSPEC’, ‘2006jan10’]
[2, ‘nir’, ‘high’, ‘NIRSPEC’, ‘2005dec11’]
[3, ‘nir’, ‘high’, ‘NIRSPEC’, ‘2001oct09’]
[4, ‘nir’, ‘med’, ‘NIRSPEC’, ‘0000xxx00’]
[5, ‘nir’, ‘low’, ‘IRTF SpeX Prism’, ‘2003sep17’]
[6, ‘nir’, ‘phot’, ‘2MASS’, [‘H’, ‘K’, ‘J’]]

From this, you can see that there are five spectra in the database: one optical-medium resolution, three nir-high resolution, one nir-medium resolution, and one nir-low resolution. You also see that there is nir-photometry for this target for the J, H, and K bands.

Note that the second row gives you the index corresponding to this target in the database. This number is the same result that you would get if you used the method bdnyc.match_unum(unum).

One last thing, to make this output code-friendly, I added the option to give you the output as a list object —as opposed to just a terminal print-out— which you can then easily manipulate with your code. To do this, type:

> x = bdnyc.show_data(‘U10276’, dump=True)

And x will hold the list result shown above. Once you can see what data exists for a target, it becomes much easier to access the actual spectrum or photometry you want.

This is the complete method:

    bdnyc.show_data(unum, dump=True)

Editing astrotools

Posted on by
Reply

It is important to maintain some standards in astrotools. The goal is to keep it readable and stable. With than in mind, I shall provide some coding guidelines and then describe the structure of astrotools in detail to allow for its easy editing.

Actually, I will let Guido van Rossum (the father of Python) provide you with all the necessary coding guidelines, from naming conventions to code lay-out. You can find them in Guido’s Style Guide for Python Code.

Let me clear out an important aspect about Python modules. A module is a collection of reusable pieces of code, such as functions and classes. A piece of code in a module can use another piece within the same module or in another module, or it can be completely independent. The implications are that you can add a function:

  • that is completely self-contained and that does not interfere with anything else in astrotools
  • that makes use of other astrotools functions/classes
  • that makes use of other Python modules (e.g. numpy), in which case you need a command line to invoke such module (more on this below).

The order of things in astrotools

The module astrotools is broken down into five sections. From top to bottom in the code, they are described below.

I) Documentation: Contains general information about the contents of the module. This type of non-executable text (a.k.a. docstring) must always start and end with three apostrophes (which I like to call Melchior, Caspar, & Balthazar). This tells Python that the stuff within them is a string that can be called by the end user using Python help.

II) External Modules: Contains the Python commands used to invoke external Python modules —the famous “import” commands. Having them all in one location as opposed to inside the functions/classes where they are needed helps to avoid repetition (when more than one function/class needs to invoke the same external module) and to clearly see the external modules needed to use astrotools.

III) Public Functions: Organized alphabetically, these functions are meant to be accessible to end users. These functions have their own docstrings, surrounded by the three apostrophes, just like in Section I.

IV) Non-Public functions: Organized alphabetically, these functions are only used by other functions/classes within astrotools. As such, end users have no business accessing these. Python convention says non-public names should be preceded by double underscore. Technically, anyone can still access them, but Python code of conduct tells end users not to do it. These functions do not need docstrings, only commented (#) intro lines to explain their purpose and where they are used.

V) Public Classes: Organized alphabetically, these classes are meant to be accessible to end users. Just like with Public Functions, Public Classes have their own docstrings.

How to add a function/class to astrotools

  1. Download the latest version of astrotools from the GitHub repository into your computer.
  2. If your function/class needs to import an external module, check if it is already imported in Section II of astrotools. If not, add the command line to do so. Take as much advantage as you can of the modules already imported in astrotools before adding a new one. More external modules needed by astrotools means more requirements that end users will have to meet before having a fully functional astrotools module.
  3. Add your function/class in the appropriate location (alphabetically) within the relevant section (III or V) of astrotools.
  4. Your function/class MUST have a docstring, and it should follow reStructuredText syntax. No panic! It is very simple. For instance, if you want a word in italics, you write: *word*. Consult the reStructuredText User Reference for more information. In terms of format and contents, follow the example of docstrings of existing astrotools functions. Include in the docstring a description of all input parameters in your function/class, including their formats. Remember, the point of documenting your code is to allow others to easily use it and edit it in case you die. This is why I like to call the docstring “the death string”.
  5. If your function/class comes with a sub-function, that is, a side function that you wrote that is used by your main function/class, this means you are a good (object-oriented) programmer! The place to put it is Section IV of astrotools. Include in it intro comments about its purpose and where it is used. If you think that this sub-function could also be useful to end users, by all means add it in Section III instead. Just remember to write a docstring for it.
  6. Test the function/class that you added. Test extreme cases, test wrong input formats, test, test, test!
  7. Update astrotools.py in the GitHub repository with your new edited version.
  8. Inform the person that maintains the public documentation for astrotools so that it gets updated in bdnyc.org/astrotools.

Intro to astrotools

Posted on by
Reply

The BDNYC Python module astrotools is now up and running for all to use and edit. This module is intended to be a collection of useful pieces of code for our everyday data handling needs. So far, astrotools is useful to handle fits files and spectral data. Additions to astrotools are welcome.

Let me clear out some relevant Python concepts. A module is simply a set of functional pieces of code. These pieces can be either functions or classes, among other things. A function is a code that executes an action, for example “make me a sandwich”. A class is a code that creates an object with specific characteristics and it allows you to handle this object very effectively. For the example above, you can have a class to create the person that will make the sandwich, a class to create a slice of mortadella, and a class to create mayonnaise. Of course, you can have functions that make no use of classes. More on that in a future post.

Where to find astrotools and how to install it

The module astrotools is available at https://github.com/BDNYC/astrotools as astrotools.py. Make sure that you download it into one of your Python folders in sys.path, which are the folders where Python looks for modules whenever you invoke one in your Python session. If you have no idea what I am referring to, please read this piece of Python documentation.

How to use astrotools

Now you are ready to invoke astrotools into your Python session. Type:

“import astrotools”

That’s it! You are ready to use astrootols functions and classes. For example, if you want to use the function that gets spectral data from a fits file, type:

“x = astrotools.read_spec(filename)”

and x will hold the output of the read_spec function. To learn about the functions and classes in astrotools, visit https://bdnyc.org/astrotools.

By convention, modules are invoked using a nickname, if only to make typing easier. I
suggest the nickname “at”. Then, import astrotools by typing:

“import astrotools as at”

To use it, type:

“x = at.read_spec(filename)”

In later posts I will describe in greater detail some important Python concepts as well as some guidelines to follow when editing astrotools.